MedPAC Recommends That CMS Streamline APM Portfolio

In its June 2021 report to Congress, the Medicare Payment Advisory Commission (MedPAC) recommended that the Centers for Medicare & Medicaid Services streamline the number of alternative payment models it deploys.

In its report, MedPAC noted that many new payment models have generated gross reductions in Medicare expenditures, indicating that APMs can successfully motivate providers to deliver care more efficiently. In addition, some models have been shown to modestly improve the quality of care.

However, the commission said that “continuing to test a large number of independent APMs may inhibit the ability of these models to reach their full potential. The commission contends the time has come for CMS to adjust its approach to designing and implementing APMs. “APMs may have a better chance of succeeding if the number of such models is reduced and the remaining models are more deliberately designed to work together to improve care quality and reduce spending, such as through more consistent model features.”

MedPAC recommends that Secretary Xavier Beccera implement a more harmonized portfolio of fewer alternative payment models that are designed to work together to support the strategic objectives of reducing spending and improving quality.

In an April 20 talk to the National Association of ACOs, Elizabeth Fowler, J.D., Ph.D., the recently named director of the Center for Medicare and Medicaid Innovation (CMMI), said the organization is at a “crossroads,” and is undergoing a strategic evaluation to determine how its alternative payment models can have a transformative impact on the healthcare delivery system. She added that CMMI will be looking at models with a health equity lens.

According to MedPAC, an important lesson of the last decade is that “implementing a large group of models that operate more or less independently of one another can have unintended consequences that dampen incentives for providers to furnish care more efficiently.”

Addressing this situation will require a change in the way Medicare approaches APM design and implementation, MedPAC said. “Instead of operating a series of APMs that are effectively developed independent of one another, the agency should seek to deploy a more parsimonious portfolio of models that are designed to work together. It is especially important to ensure that financial incentives presented by different models are complementary and do not weaken one another when combined.”

One way to accomplish this would be to focus on a single population-based model with different tracks by provider type or beneficiary population, the report said. For instance, there could be separate, but aligned, tracks for integrated health systems, multispecialty physician practices, end-stage renal disease facilities, and so on. Other types of models, such as those that focus on episodes of care or primary care transformation, could be added to the portfolio to act as an extension of the main population-based model, although model overlap rules would need to carefully consider how best to incentivize optimal management of beneficiaries treated by two sets of providers in two different models.

Accounting for interaction between an ACO and an episode-based payment model is especially important, since both models can hold participants accountable for the cost of care of a common set of beneficiaries during the same period of time, the report notes.

A second approach would be to take a geographic approach to testing models, which CMMI has done for some models such as CPC+). CMMI could limit all of its models to particular geographic areas of the country, to more actively control how many models are operating in any given region at once. For instance, certain geographic regions could have access to the MSSP only, with no other Medicare APMs operating in those areas. This approach could reduce the potential for patients to be attributed to multiple models (although it would not eliminate this problem) and could allow researchers to identify the additive impact of coupling certain models compared with implementing some models by themselves.

One benefit to reducing the number of APMs is that it would make the task of standardizing model parameters a more manageable undertaking for CMS. If models were less complex, they could also attract more independent providers, since such providers might no longer need to hire consultants to help them understand different models, enroll in a model, and excel in that model, MedPAC said.

Still a third approach that could be contemplated would be to encourage more states to follow Maryland and Vermont’s lead by pursuing waivers that allow them to operate a smaller set of state-specific versions of CMMI’s APMs within their borders. Maryland couples its unique global payment model for hospitals with state-specific versions of BPCI Advanced and CPC+ and an additional state-specific model that lets hospitals design their own payment incentives for providers in their communities. Vermont is operating only a state-specific version of the Next Generation ACO Model.

MedPAC says CMS could work with other states to implement different combinations of customized versions of its payment models in an effort to identify the combination of models that will best engage the widest range of providers to produce the largest impacts on spending and quality.