MedPAC Continues Discussion on Unified Prospective Payment System for Post-Acute Care
January 22, 2016 11:58 AM
On Friday, January 15, 2016, the Medicare Payment Advisory Council (MedPAC) held a meeting to continue its discussions on developing a unified prospective payment system (PPS) spanning the post-acute care (PAC) settings. As mandated under the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014, MedPAC is required to develop a prototype PPS, using the uniform assessment data gathered previously during the Centers for Medicare & Medicaid Services (CMS) Post-Acute Care Payment Reform Demonstration (PAC-PRD) completed in 2011.
Current policy involves four separate, setting-specific payment systems with different payments for similar patients. MedPAC has been critical of the home health and SNF payment systems based on the perception that they encourage providers to furnish therapy that is unrelated to a patient’s care needs. The objective of the unified payment system would be to span the four settings and correct perceived short-comings, basing payments on patient characteristics and not site of care. The first of the two mandated reports is due at the end of June 2016, and it must recommend features of a unified payment system and, to the extent feasible, estimate the impacts of moving to such a system. A second report, due around June 2023, must propose a prototype design.
This was the third session MedPAC has held considering the design, testing and impacts of a PAC PPS. At the September meeting, MedPAC discussed the overall approach to the mandate and the results of modeling stays in CMS’s PAC demonstration, and determined that a unified PAC PPS looked feasible. At the November session, MedPAC discussed possible complementary policies that could be implemented to counter volume incentives that would remain in the payment system, including a readmission policy, value-based purchasing program, and the use of a third-party PAC benefit manager. MedPAC also discussed waiving setting-specific regulations and moving toward a common set of conditions of participation.
During the January 2016 meeting, MedPAC discussed the results of a modeling analysis of PAC stays in 2013 under a unified payment system; considered the need for certain payment adjusters; and estimated the impacts on payments. The mandated report due in June 2016 has two requirements: 1) Evaluate and recommend features of a PAC PPS using data from the PAC-PRD; 2) Consider the impact of implementing a unified PAC PPS. In order to comply with the mandates, MedPAC developed three models, each with a specific purpose. The first model uses unique data in the PAC-PRD to test the feasibility of a PAC PPS. The second “administrative” model predicts relative costs of PAC-PRD stays, without relying on unique PAC-PRD data, and compares the accuracy of models using same stays, in order to assess the accuracy of an administrative model that could be used on a large sample of stays. The third “administrative” model estimates impacts with all PAC stays. MedPAC relied of a variety of patient groups to evaluate the results, including clinical groups, a variety of impairment and severity groups, community groups, and others such as the aged and disabled.
In testing the models, MedPAC found the full and administrative models predicted “very similar” relative costs of stays for most groups and explained similar shares of the variation in costs across stays. MedPAC also found that the administrative data can be used to establish accurate relative weights for most groups and estimate the impacts of a PAC PPS. However, MedPAC found certain groups had average predicted costs that deviated from average actual costs, and which “may warrant payment adjustment,” including unusually short stays (to prevent large overpayments), and high-cost outliers (to protect providers from large losses). MedPAC also found differences that “may warrant further study” including low-volume, isolated providers (to ensure access) and extremely sick patients (to ensure access).
With regards to estimating the impact of a PAC PPS, MedPAC assumed the same level of aggregate payments as under current law; did not reflect policy changes since 2013; and did not assume any changes to provider behavior. MedPAC found that a PAC PPS would narrow the difference between payments and actual costs across most patient groups. A PAC PPS is estimated to shift payments across stays, with payments increasing for some groups and decreasing for others. For example, payments would increase for many of the medically complex and patient impairment and severity groups. Meanwhile, payments would decrease for clinical groups where rehabilitation therapy is a key component of care with therapy services unrelated to patient characteristics; and also decrease for many types of stays treated in higher-cost settings that are also treated in lower-cost settings.
Among the implications from the analysis was that payments for stays in home health will need to be aligned with the setting’s lower cost. In addition, MedPAC stated that a payment adjuster for short-stays will likely be needed, and that low-volume, isolated providers may need protections.
MedPAC members generally agreed that a PAC PPS is feasible and would break down silos between settings. In March 2016, MedPAC will meet again to further discuss the need for an adjuster for low-volume, isolated providers and look at a prototype outlier policy. MedPAC will also discuss different ways to establish the aggregate level of payments. In April 2016, MedPAC will finalize the report, which is due at the end of June 2016.
The National Association for Home Care & Hospice(NAHC) is open-minded with regards to the development of a unified PAC PPS. NAHC awaits further detail on the prototype, as the lack of detail that MedPAC has provided so far prevents a full assessment of its potential. NAHC’s position is that any such effort must:
Take into account that the majority of home health services do not meet the definition of post-acute care, based on the fact that many patients enter home health from the community rather than from an institutional setting;
Consider that, whereas all of the other relevant settings are under the control of the provider, the home is under the control of the patient, which creates different issues with regards to risk adjustment;
Recognize that home care for many patients is a clinically necessary setting, not just an acceptable one, based on the patients’ condition and the fact that they are protected by being at home;
Be developed with full transparency and fully tested for reliability.
Stay tuned to NAHC Report for further analysis of MedPAC’s efforts on a unified PAC PPS.