Functional-based risk adjustment for Medicaid Managed Long Term Services and Supports

Establishment of Medicaid Managed Long Term Services and Supports (MLTSS) programs has been increasing in frequency as both the number of nursing home eligibles and their corresponding service costs rise, especially as the Baby Boom generation enters this phase of their lives. Payments made to managed care organizations (MCOs) in these programs are often adjusted for the location of care of their members (e.g., nursing home versus home- or community-based). This is a blunt adjustment mechanism that can result in payments not well matched to underlying member risk levels in some situations. An increasing nationwide initiative to develop tools to adjust MCO payments using the functional status of their members has the potential to better match payment to risk and to better align MCO and program incentives. Such functional-based risk adjustment (FBRA) adjusts MCO capitation rates based on member functional capabilities, e.g., requiring assistance with activities of daily living (ADLs) as well as behavioral, medical, and/or demographic characteristics.

MLTSS payment structure encourages appropriate community-based treatment

A primary objective of MLTSS programs is to improve member quality of life and reduce program costs by transitioning members from nursing homes to less costly home- or community-based care (referred to as “community-based” care in the remainder of this paper), or by delaying admission into nursing homes as long as feasible. Medical necessity criteria and quality of care metrics help to ensure that appropriate care is provided to members regardless of the location of care. An MCO capitation payment structure based on location of care provides some financial incentives consistent with the primary program objectives of reduced cost and improved quality through community-based treatment.

Many MLTSS programs adjust MCO capitation for variation in the location of care mix of their members.¹ MCOs enrolling a greater-than-average percentage of members in nursing homes will receive higher capitation and vice versa. The calculation of the MCO location of care mix is then typically updated infrequently, often once a year, in order to introduce a financial incentive to MCOs to provide services in the community at less cost than a nursing home. If an MCO increases its percentage of members treated in the community over the course of the contract period, the MCO will realize a financial gain. In subsequent contract periods, the MLTSS program will realize the cost savings associated with increased community-based treatment. Figure 1 illustrates how an MCO could realize a financial gain of $80 per member per month (PMPM) by reducing the utilization of nursing home services relative to the assumption used during capitation development.²

Location-based rate structure payments are sometimes not consistent with program goals

While a location of care-based rate structure generally encourages MCOs to enable appropriate community based treatment, in some situations MCO payments are not consistent with program goals for MCO initiatives. Following are three examples of MCO payment results under a rate structure based on location of care that are typically not consistent with program goals.

Example #1: No variation in payments for differing levels of member risk

The location of care rate structure typically does not vary capitation payments for members treated within the same location of care even though their care needs may vary significantly. This may result in a less optimal allocation of capitation among MCOs with varying levels of member needs than would occur with a more robust risk adjustment mechanism.

Example #2: Member transitions into the community late in a contract period

If the location of care mix is calculated at the beginning of each contract period, it might be detrimental to an MCO’s financial interest to realize a member’s discharge from a nursing home into community-based treatment near the end of the contract even though timely transitions are a typical MLTSS program goal. The MCO bears costs associated with enabling community-based treatment, and an MLTSS program should include sufficient financial incentive to offset those costs. A discharge happening near the end of a contract, however, may not leave enough time to realize sufficient financial gains before the member is reclassified as a community-based member, with the associated lower capitation rate, at the beginning of the next contract. Figure 2 shows a simplified illustration of the scenario over two contract periods for a November member transition with a calendar year contract renewal and location of care mix calculation.

If the same discharge instead occurs early in the next contract period, the member is identified as a nursing home member in the mix calculation, thereby increasing capitation for the second contract period. Then the MCO will realize financial gains for most of the second contract period once the member is discharged from the nursing home. Figure 3 shows a simplified illustration of the scenario where the transition occurs in February of the second contract period. Note the increase in total MCO gain from $8,000 in Figure 2 to $44,000 in Figure 3.

Example #3: Transitions into the community for high-need members

As mentioned previously, there is a wide range of service needs and costs for members receiving LTSS in the community depending on their functional capabilities and other characteristics. Continuing the illustration from Figure 1, consider a member receiving care in a nursing home who would require $4,000 in monthly services if treated in the community. Transitioning the member to the community would ultimately generate $1,000 ($5,000–$4,000) in monthly program savings. However, the MCO would realize $3,000 ($4,000–$1,000) in monthly losses under community-based treatment because the capitation rate for all community-based members is fixed. Assuming the deviation in actual nursing home costs versus those assumed in capitation is less than $3,000, the MCO incurs a financial penalty when transitioning the member into the community. This result conflicts with typical MLTSS program goals.

FBRA better matches MCO payments to member risk and program goals

Adjusting MCO capitation rates based on member functional capabilities and behavioral, medical, and/or demographic characteristics addresses the concerns in the previous section. Medicaid programs in both Wisconsin and New York have implemented such FBRA models for their MCO capitation payments. These models do not give consideration to the location of the care provided. Rather, they allocate capitation to MCOs based on the average, incremental program cost associated with a particular member characteristic, including both nursing facility and community-based costs.

While there are significant data and process hurdles to clear before implementing FBRA, there are also significant program improvements. Following are the ways FBRA addresses the concerns from the previous section:

Example #1: No variation in payments for differing levels of member risk

Unlike rate structure based on location of care, FBRA recognizes differences in member service needs through different MCO capitation levels beyond simply recognizing nursing home placement. The FBRA model in Wisconsin is realizing predictive “R-squared” metrics of 35% to 50%, which are significantly higher than metrics typically seen for acute care risk adjustment models.

Example #2: Member transitions into the community late in a contract period

Because MCO capitation is calculated using member characteristics rather than location of care, it is in the MCO’s financial interest to improve the efficiency and quality of LTSS delivery regardless of the current location of care. Transitioning a member from a nursing home to the community will not directly influence the capitation associated with that member, so MCOs retain the incentive to transition members from nursing homes to the community as quickly as is appropriate.

Example #3: Transitions into the community for high-need members

Because changing the location of care for members does not directly influence capitation rates, there is a financial incentive for MCOs to encourage delivery of care in the most efficient manner for each member regardless of the member’s service needs or current location of care. MCO capitation under FBRA will be higher for members with high levels of service needs, and MCOs that achieve any gain in overall service efficiency will realize improved financial results. In Example 3, if an FBRA model generated a capitation rate of more than $4,000 for the high-need member, the MCO would realize a financial gain rather than a penalty when transitioning the member from the nursing home into the community.

FBRA can also improve Medicaid agency budgeting and MCO care management models

Implementing FBRA in an MLTSS program requires significant policy, process, and calculation efforts. Perhaps the largest hurdle to clear for most states is the consistent use and reporting of a common functional assessment tool by a state and its MCOs for all program members. The second paper in this series will address some of the implementation challenges. However, even if a program is not at a place where FBRA can be implemented for capitation rate adjustment, there are still ways for MLTSS programs and MCOs to benefit from FBRA.

MLTSS program budget projection

In most current MLTSS programs, MCOs do not administer and report results for the same, state administered functional assessment tool after the initial member assessment upon entry into the program. Rather, MCOs typically develop custom screening tools to better match the structure of their care management activities. However, that initial state-administered screen can still be valuable to state Medicaid program multiyear budget forecasts. For this application, rather than developing an FBRA model for a one-year period, the model projects successive years of costs for a member from the member’s functional assessment upon entry into the program. If sufficiently complete and accurate fee-for-service or encounter data is available for the program, multiple years of this experience and initial member assessments can be used to build a longitudinal cost model to better inform budget expectations in future years.

MCO care management efficiency

MCO resources available for care management activities are limited, so it is important to prioritize those resources for members whose LTSS costs and quality of life may be most impactable. MCOs often utilize custom member screening tools to determine the level of care manager activity and approved LTSS hours for each of their members. FBRA models, either developed by the MLTSS program or by an MCO, can identify inconsistencies between the care manager activity and/or approved LTSS hours and the modeled costs for particular conditions. Members with high FBRA-modeled costs and low approved service hours, or vice versa, may identify opportunities to review and improve care management policies.

Efforts to incorporate and improve quality of FBRA are increasing

As the number of MLTSS programs increases, there is also increased realization of the value of functional-based risk adjustment of MCO capitation. Two MLTSS programs have implemented FBRA, and at least half a dozen others are exploring other uses for FBRA and/or the potential for implementation in the future. While there can be significant administrative changes required to utilize FBRA, the improvements in program financial incentives and member quality of life can be just as significant.

¹ While there are a variety of rate structures across MLTSS programs based on location of care, for purposes of this white paper a program structure with two locations of care is considered: nursing facility versus all other community-based care.

² For simplicity’s sake, it is assumed 1) that the starting capitation rate equals the expected LTSS cost, ignoring the impact of MCO administrative costs and other retention loads, and 2) that members transitioning from a nursing home into the community have the same LTSS costs as the average community member.