Download as PDF

NAACOS Summary of Key Elements of the June 2021 MedPAC Report

Overview: The Medicare Payment Advisory Commission (MedPAC) is an independent congressional agency that advises Congress on issues affecting Medicare, primarily payments to providers in the traditional fee-for-service (FFS) program and private health plans participating in Medicare. Although formally advising Congress, the Administration, including the Centers for Medicare & Medicaid Services (CMS), pays close attention to MedPAC’s recommendations as well. The 17-member commission meets monthly between September and April and publishes two reports a year: March and June. MedPAC’s June 2021 report includes a chapter dedicated to alternative payment models (APMs), highlighting successes of ACOs. The full report is available here, and MedPAC’s press release is here. MedPAC also devoted a great deal of attention to ACOs in its June 2020 report, which NAACOS summarized here, as well a full chapter on ACOs in the June 2019 report, which is summarized here. 

NAACOS is pleased to see the MedPAC continue to dedicate attention to the move to APMs. MedPAC highlights the value and successes of the Medicare Shared Savings Program (MSSP) and other ACO models tested through the Center for Medicare and Medicaid Innovation (CMMI). MSSP has been Medicare’s largest and most successful payment model and will continue to play a critical role in Medicare’s transition away from FFS payment and to a system based on value. We are pleased to see MedPAC acknowledge that population-based models such as ACOs have generated the most consistently favorable results for the Medicare program. NAACOS encourages MedPAC and policymakers to provide ACO investment opportunities, give deference to ACOs when model overlap occurs, and test innovations within the ACO model to further the transition to value. 

Key MedPAC recommendations on APMs included:

Summary of Key MedPAC Report Sections Related to Value

“Streamlining CMS’s portfolio of alternative payment models”
Chapter 2 of the MedPAC report discusses recommendations on the future of CMS’s APMs. In line with what we have heard from CMMI Director Liz Fowler, MedPAC recommends that CMS refine and harmonize its portfolio of APMs. In 2021, CMS will operate 12 APMs with 25 tracks and has operated 54 models in the past decade.

MedPAC summarizes existing evidence about Medicare APMs, saying population-based models such as ACOs have generated the most consistently favorable financial results for the program. The most consistent results by ACOs were reduced inpatient and emergency department use and increased delivery of preventive services and chronic disease management. Episode-based models have also generated savings, though less consistently than ACOs, and reduced the rate of readmission for hip- and knee-replacement patients. Primary care transformation models, such as Comprehensive Primary Care Plus, have produced inconsistent results, often with financial losses and little to no effect on quality. They note when providers are held accountable for total cost of care, such as in ACOs, providers have more flexibility in how they deliver care and may generate savings. See the excerpt of Table 2-1 below with results of select models. 

Excerpt of Table 2-1: Evaluation findings for CMS’s key Medicare APMs

CMMI model	Years operated	Beneficiaries/ episodes	Gross savings/ losses	Net savings/ losses	Impacts on quality
Population-based models (ACOs)
Physician Group Practice Demonstration	2005-2010	221,000	$171 PBPY savings	$69 PBPY savings	Reduced hospital admissions and ED visits
Pioneer ACO Model	2012-2016	608,000	$427 PBPY savings	$316 PBPY savings	Improved hospital admissions for COPD, and physician follow-up
Next Generation ACO Model	2016-2021	1,399,000	$112 PBPY savings	$38 PBPY losses	No significant change
Models that facilitate participation in population-based models (ACOs)
Advance Payment ACO Model	2012-2015	284,000	$14 million savings	$87 million losses	No significant change
ACO Investment Model	2015-2018	447,000	$339 PBPY savings in 1st year	$280 PBPY savings in 1st year	Reduced hospitalization, ED visits, and SNF care
Episode-based payment models
BPCI Model 2	2013-2018	1,260,000 episodes	$947 savings per episode	$332 losses per episode	No significant change
BPCI Advanced Model	2018-2023	208,000 episodes first 10 months	$646 savings per episode	$761 losses per episode	Mortality increased slightly
Comprehensive Care for Joint Replacement Model	2016–2024 (first 3 years evaluated for mandatory hospitals)	115,000 episodes over 3 years	Savings $1,323 per episode (5%)	Savings $536 per episode (2%)	Reduced rates of unplanned readmissions and certain complications
Primary care transformation models
Comprehensive Primary Care Plus Model	2017–2021 (first 3 years evaluated)	1,900,000 beneficiaries	Losses $36 PBPY (0.3%) (Track 1); $19 PBPY (0.2%) (Track 2)	Losses $198 PBPY (2%) (Track 1); $313 PBPY (3%) (Track 2)	Slight decreases in ED visits; slight increases in diabetes services, breast cancer screenings, and follow-up after hospitalization

MedPAC discusses several factors that may be limiting the success of APMs. Because providers in APMs may still operate under traditional FFS payment, there are remaining incentives to increase the volume of services. Additionally, APM incentives can be more complex than traditional FFS incentives, with significant upfront investments of time and resources which may suppress provider participation or limit the effectiveness of incentives for providers to change their behavior. When incentives do not flow directly to clinicians, this may also reduce their incentive to change behavior. Another issue they note is that APMs may take more time to generate savings and improve quality than CMMI currently allows. Evidence shows that it takes years for providers to adopt best practices into regular workflows and therefore models tested for longer periods of time may show greater success. When payment models are voluntary, selection bias among participants may occur, thus limiting the impact of the model. They also point out that the upfront infrastructure investments required to succeed in new APMs may be unattainable for some clinicians and that beneficiaries’ incentives are not always aligned with those of providers. MedPAC recommends CMS consider these challenges that contribute to the underperformance of models. 

MedPAC describes several challenges and unintended consequences of implementing concurrent APMs, including dilution of model incentives by allowing providers to participate in multiple models and attributing beneficiaries to multiple models. Not only does this weaken the programs’ incentives, but it can also inhibit evaluators’ efforts to accurately assess the impact of a given APM. In order to address these concerns, MedPAC recommends CMS implement a more harmonized portfolio of fewer APMs that are designed to work together to support the strategic objectives of reducing spending and improving quality. 

MedPAC describes several different ways this recommendation could be implemented, including to focus on a single population-based model with different tracks by provider type or beneficiary population. Other types of models such as episode-based or primary care transformation could be added as extensions within the population-based model. They also suggest taking a geographic approach, limiting all models to particular geographic regions to control how many models are operating in a given region at once. For either of these options, they recommend CMS utilize standardized model parameters to reduce complexity and attract more independent providers. The third approach they recommend is to encourage states to pursue waivers that allow them to operate a smaller set of state-specific models, such as Vermont’s multi-payer ACO model. 

“Rebalancing Medicare Advantage benchmark policy”
MedPAC recommends several changes to the way benchmarks are determined for the Medicare Advantage (MA) program. They note issues with the current quartile system, which was enacted by the Affordable Care Act (ACA) to ensure broad access to MA plans in low-spending areas by setting higher benchmarks. MedPAC states benchmarks in areas with the lowest FFS spending are likely higher than necessary to induce plan participation. This has led to a disproportionate share of MA enrollees in areas with the lowest FFS spending. Another issue with the quartile system is that small differences in FFS spending can cause large differences in benchmarks, resulting in instability in payment rates over time. 

Current estimates show Medicare spends 4 percent more per capita for beneficiaries in MA than those enrolled in traditional FFS Medicare, and a recent decline in plan bids to levels below FFS spending suggests that Medicare could share in plan efficiencies by reducing payment benchmarks. MedPAC recommends that Congress implement a new MA benchmark policy that uses a relatively equal blend of per capita local area FFS spending and standardized national FFS spending in order to allow the program to capture additional efficiencies where plan bids are lowest relative to their benchmarks. Additionally, they recommend a rebate of at least 75 percent and a discount of at least 2 percent in order to incentivize plan efficiency and ensure Medicare shares in the efficiencies generated. MedPAC also recommends applying their prior MA benchmark recommendations to use geographic markets as payment areas, using the FFS population with both Part A and Part B in benchmarks, and to eliminate the current pre-ACA cap on benchmarks. MedPAC asserts that implementing these changes could result in about two percentage points in savings to the Medicare program. 

“Evaluating the skilled nursing facility (SNF) value-based purchasing program (VBP)”
MedPAC describes fundamental design flaws in the SNF VBP program and thus recommends the program be terminated and replaced. They found five major shortcomings of the current program, including the use of a single outcome measure. They also say that the minimum stay counts are too low to ensure that the program rewards performance rather than random variation, and performance scoring “cliffs” may not provide enough encouragement for improvement. Lastly, the program does not distribute the pool of incentive payments but retains a portion as program savings. While Congress attempted to address some of these issues with the Consolidated Appropriations Act of 2021, fundamental flaws remain. 

MedPAC recommends replacing the SNF VBP program with a new value incentive program and that CMS should require SNFs report patient experience measures. MedPAC also suggests that public reporting of provider performance could hold providers accountable to consumers and encourage improvement. 

“Private equity and Medicare”
Congress requested MedPAC look at the role that private equity (PE) plays in Medicare. MedPAC notes that transparency of ownership could help beneficiaries select providers, and researchers analyze the effects of PE backing. CMS collects provider ownership data for the enrollment process, payment, and fraud prevention, but the data submitted to CMS are often incomplete or inaccurate. For hospitals, PE-owned facilities had lower costs and lower patient satisfaction. For physician practices, there was little evidence available on the impact of private equity ownership. MedPAC finds that PE funds own about 2 percent of MA companies as of January 2021. Many PE firms have invested in a range of companies that work for MA plans such as care management services. They did not find any research that examines the effects of PE investments in MA companies on Medicare costs. 

“Medicare beneficiaries’ access to care in rural areas”
MedPAC updated its June 2012 report analyzing access to care among rural beneficiaries by comparing their use of services with that of urban beneficiaries. The June 2012 report also established principles to guide policies with respect to rural access to, quality of, and payment of care. The results of the updated report find that rural and urban beneficiaries have similar access to care, though there are some differences, including less outpatient specialist use by rural beneficiaries, and greater use of hospital outpatient services. MedPAC plans to expand this utilization analysis to include information on duals, those with multiple chronic conditions, or those residing in a medically underserved area with a report to be published June 2022.

Conclusion

NAACOS agrees with MedPAC’s recommendation to focus on population-based models, such as ACOs, which are accountable for the total cost of care of assigned populations. Testing new innovations as expansions within MSSP provides a unique opportunity to explore new payment and delivery options while controlling for costs and quality.