MedPAC Backs Simplifying Medicare Alternative Payment ModelsJune 16, 2022
WASHINGTON — The Medicare Payment Advisory Commission (MedPAC) proposed streamlining Medicare alternative payment models (APMs), along with policy options for tackling pricey Part B drugs, in its June report to Congress Wednesday.
In the June 2021 report, MedPAC called for reducing the number of Medicare APMs so that models running concurrently could work better together. The 2022 report builds on that idea, offering strategies that “would represent a shift for CMS — moving away from temporarily testing a large number of model tracks on a small scale to permanently operating a small number of model tracks on a large scale.”
During a press call, James Mathews, PhD, executive director of MedPAC, stressed “that the Medicare alternative payment model landscape should be centered around a single population-based model — think ACOs [Accountable Care Organizations] — with a small number of easy-to-understand tracks embedded within that model that providers could participate in based on their ability to take on risk for a given population.”
CMS oversees a broad range of APMs with multiple tracks of the Medicare Shared Savings Program, such as the ACO Realizing Equity, Access, and Community Health Model, which has two tracks, and various episode-based payment models, such as the Comprehensive Care for Joint Replacement Model and the Bundled Payments for Care Improvement Advanced Model.
To address the “unnecessary complexity” of having so many different models, which can potentially “dilute incentives” from one to the next, MedPAC suggested centering all models around a single population-based model, with three tracks based on size of the provider groups and willingness to accept financial risk:
- Track 1: Target small provider groups — such as independent primary care practices that join together to form an ACO — and offer the possibility of “modest shared savings,” without the responsibility of repaying shared losses
- Track 2: Focus on midsize provider groups with the potential to earn a higher percent of shared savings and take on risk for shared losses
- Track 3: Suitable for large provider groups that would be held accountable for full risk for all Part A and Part B spending for their beneficiaries
A second option would be to create a single-track population-based payment model, where the shared savings and loss rates differ based on the characteristics of the ACO, including its ability to assume financial risk. Under this single population-based model, the Commission suggested “rebasing” ACO spending benchmarks. Currently, benchmarks are updated every 3 years based on an ACO’s “actual recent spending,” according to the report, which leads to participants competing against themselves, and even penalizes their success with “harder-to-beat benchmarks.”
If an ACO can no longer reach its benchmarks, “it may create an incentive for ACOs to abandon the program,” Mathews explained.
Under the single model, benchmarks would be based on administrative updates instead of actual spending, he said. Historical spending could be “trended forward” to the current year using a growth factor, such as a price component or the volume and intensity of the service, the report noted. As a result, ACO participants would have a “clear and predictable” sense of the targets they need to shoot for in a given year, according to MedPAC.
The report highlights the need for mandatory episode-based models “that have been demonstrated to produce positive results in terms of savings or outcomes,” Mathews said. “Any beneficiaries who experienced a diagnosis that corresponded to one of these episodes, would be episode participants by default, and their care would be provided by episode participating providers.”
However, MedPAC did not cast the population-based or ACO models as mandatory in the immediate future, given that not all participants are ready to accept downside risks. However, the Commission encouraged the CMS Innovation Center to continue testing episode-based payment models with the goal of identifying episode types that could be added to the model in the future.
To prevent models from working against each other with conflicting incentives, “we would allocate any bonus payments so that episode participants would have a strong incentive to deliver care efficiently, that ACO participants would have an incentive to use low-cost, high-quality episode providers. And then lastly, we would assert that the total bonus payments made to each of the sets of participants should not result in increased Medicare spending,” Mathews explained.
Part B Drugs
Spending on Medicare Part B drugs, biologics, and other physician-administered medications increased by an average rate of close to 10% annually from 2009 to 2019. While spending ticked up around 4% in 2020 — likely due to reduced healthcare utilization during the pandemic — price concerns have not waned.
Medicare pays for most Part B drugs at a rate of the average sales price (ASP) plus 6% (106% in total). Currently, the program lacks the right levers to pay for these drugs in a way that appropriately captures different competing factors, such as the drug’s net clinical benefit, the manufacturer’s effort and innovation, and patients’ ability to afford the medications.
MedPAC proposed three strategies for addressing different aspects of the Part B drug space.
For “first-in-class” drugs, Medicare currently lacks the authority to weigh a new Part B drug’s net clinical benefit against the standard of care when setting payment rates. As a result, such rates can exceed a new drug’s clinical effectiveness.
To address those first-in-class drugs that have been approved with “uncertain clinical evidence” — for example, on surrogate or intermediate clinical endpoints through the FDA accelerated approval pathway — the Commission recommended two possible policy approaches, which Congress could give the HHS Secretary the authority to apply.
First, CMS could apply “coverage with evidence development” to the product, requiring the collection of additional clinical evidence about the new drug while allowing patients to access it. Such a process should be “clear,” “transparent,” and “predictable” and provide a medium for public input, the Commission said, adding that having a standard funding stream “might ease implementation.”
The Commission also suggested placing a cap on the drug’s payment rate based on information about the product’s “estimated net clinical benefit and cost” when measured against the current standard of care. Again, such a mechanism would benefit from a “clear and predictable decision-making framework” that enables transparency and public feedback, according to the report.
Separately, to increase competition among drugs with therapeutic alternatives, the Commission recommended “internal reference pricing” or “consolidated billing,” which involves Medicare issuing “a single reference price for drugs that have similar health effects based on the Part B drug payment rates of the products in the reference group.” Consolidated billing differs from internal reference pricing, in that the former calls for assigning clinically similar products to one billing code, whereas the latter allows products to stay in their own billing code.
“Under reference pricing policies for Part B drugs, manufacturers would have incentive to lower their prices relative to competitors to make their products more attractive to providers and garner market share, which would result in savings for beneficiaries and taxpayers,” the report noted.
Lastly, to correct certain incentives under the ASP system — Medicare’s 6% add-on to ASP implicitly rewards providers who choose higher-priced drugs over lower-priced drugs — MedPAC suggested putting a fixed-dollar limit on the add-on payment or replacing some portion of the add-on with a fixed fee, or combining these approaches. Placing a fixed-dollar limit on the add-on payment would mean very expensive drugs would become less pricey, while applying a fixed fee would raise payments for some inexpensive drugs but reduce payments for more expensive ones.