REGULATORY REVIEW: CMMI: Pursuing Innovation in Medicare and Medicaid

Regulatory Review

CMMI: Pursuing Innovation in Medicare and Medicaid

A brief history

Regulatory Review

CMMI has been the subject of intense scrutiny from the health care community, a battleground for policy innovations related to care delivery and payment reform.

Regulatory Review

CMMI: Pursuing Innovation in Medicare and Medicaid

A brief history

In March 2010, after more than a year of development, President Obama signed into law the Patient Protection and Affordable Care Act (PPACA). The 900-page law represented the most significant overhaul of the US health care system since the 1960s, directing billions of dollars toward a plethora of new health care programs. Of particular note was Part III, Sec. 3021, which created a new Center for Medicare & Medicaid Innovation (referred to as CMI in the bill, now commonly known as CMMI) within the larger Centers for Medicare & Medicaid Services (CMS). In the ensuing years, CMMI has been the subject of intense scrutiny from the health care community, a battleground for policy innovations related to care delivery and payment reform. With the Trump administration signaling renewed commitment to the center, even amidst budget uncertainty, it seems worthwhile to review CMMI’s activity, and where it might go in the future.


To read this article, you have to be a member of ASCA or subscribe to ASC Focus magazine.

Already a member or subscriber? Sign in.

To become a member, click here. To subscribe to ASC Focus, click here.

CMMI has not yet engaged with any grants involving ASCs, though the ASC community has signaled willingness in recent years to engage in new, innovative payment models. Bundling payments, particularly for joint replacement procedures, is seen as good candidate for a CMMI-type test model.

Beginnings: A 10-Year Mandate for Innovation

The PPACA laid some basic logistical guidelines to get CMMI off the ground and running. The expressed purpose of CMMI was to “test innovative payment and service delivery models to reduce program expenditures . . . while preserving or enhancing the quality of care furnished.” Though the mandate was intentionally broad—the Secretary of Health and Human Services (HHS) was meant to have great discretionary power to select payment and care models to test—the law did delineate 18 model examples as guidance. These ranged from general (promote care coordination between providers and suppliers) to more specific (varying payment methodologies for advanced diagnostic imaging services). The center would begin carrying out duties no later than January 1, 2011, and was funded by $10 billion federal appropriation intended to renew every decade. Importantly, this meant that test models selected would have no budget neutrality requirement, although expansion of a model past the test phase would require the CMS actuary to show no increase in net spending.

CMS Administrator Don Berwick had appointed Richard Gilfillan, MD, acting director of CMMI, and by March 2012 the new sub-agency had almost 200 staff members. The rapid staffing matched early programmatic implementation; the first year saw 12 initiatives launched, some of which, including the Bundled Payment for Care Improvement (BPCI) and Comprehensive Primary Care (CPC) Initiative, are still ongoing today.

Transitions: Expansion from Voluntary to Mandatory Models

In June 2013, CMS Chief Medical Officer Patrick Conway succeeded Gilfillan as director of CMMI, a post he would serve until the end of the Obama administration in 2017.

CMMI would get swept into more change in June 2015 with the passing of the Medicare Access and CHIP Reauthorization Act (MACRA). The legislation, meant to reform the way in which physicians are paid under Medicare, provided incentives for physicians to join alternative payment models (APMs), of which CMMI had been the primary ground for testing and development. At that time, CMMI was running roughly 60 APM models in a number of different categories.

Meanwhile, early test models were beginning to show promising returns, most notably the Pioneer Accountable Care Organization (ACO) Model, which was the first to meet the PPACA expansion criteria (no net spending and no reduction in quality or access of care). In January 2015, Department of Health and Human Services (HHS) Secretary Sylvia Burwell set new goals for CMMI: 30 percent of Medicare payments delivered through APMs by the end of 2016 and 50 percent by 2018. Though seen as ambitious, Conway announced that the 2016 goal had been met almost a year ahead of schedule in March 2016. This coincided with the rollout of CMMI’s first mandatory model, the Comprehensive Care for Joint Replacement (CJR) model, which made roughly 800 hospitals responsible for patient recovery until 90 days postdischarge for hip and knee replacement “episodes of care.” Although viewed apprehensively, most industry stakeholders viewed CJR as an indicator of mandatory models to come.

By December 2016, in its final report to Congress under the Obama administration, CMMI reported that 18 million individuals had been impacted by, received care, or would soon receive care from a provider participating in a CMMI model.

An Uncertain Future Under the New Administration

The arrival of the Trump Administration in February 2017 brought significant uncertainty to CMMI and HHS as a whole. Although the health care industry had been generally supportive of the investing in innovative care and payment delivery models, many were troubled by the trend of mandatory models, seeing it as on overstepping of CMMI’s statutory authority. Incoming HHS Secretary Tom Price had been particularly critical of mandatory pilots during his time in Congress, asserting that these models were the government dictating to physicians how to practice. With the new administration promising rollbacks in Obama-era health reforms and overall spending, the $10 billion per decade CMMI budget seemed like a logical program for the chopping block.

Upon taking office, Price acted quickly, rolling back impending mandatory orthopedic and cardiac bundled payment programs. In September 2017, new CMS Administrator Seema Verma posted an op-ed in the Wall Street Journal announcing a new direction for CMMI. The administration planned to re-orient CMMI around provider flexibility, free market principles and consumer choice and solicited stakeholder feedback through a request for information (RFI). Later that same month, however, Price was forced to resign, replaced by pharmaceutical executive Alex Azar. In his nomination hearings, Azar struck a decidedly different tone in relation to CMMI, asserting that models may have to be mandatory in some instances to garner adequate data.

In April 2018, Azar appointed Landmark Health CEO Adam Boehler as CMMI director. The two officials have signaled that the center will become more active in both soliciting and managing test models, prioritizing those that will have immediate impact on health care consumers. In particular, CMMI will likely have a role in revising rules around Medicare Accountable Care Organizations (ACOs) and a new, somewhat vague “direct provider contracting” model. This is likely, despite continued discussion around larger administration spending cuts including some $800 million in CMMI budget cuts for 2018 and 2019 (CMMI will receive a new mandatory appropriation renewal in 2020). As of the writing of this article CMMI has 39 active innovation models and five more under development.


Despite more than seven years of operation, the evidence from CMMI is mixed at best. Some of this is intentional; many models have only been in operation for a few years and are meant to take four to seven years before being reviewed for expansion. Most models have shown some degree of cost savings, although few have seen enough savings to offset commensurate management or administrative spending increases. Proponents will say that this is just an indication that models take time to refine as payers and providers realign practice and administration along value-based lines.

From a budgetary standpoint, the Congressional Budget Office (CBO) has consistently projected that CMMI will generate net savings in the long term, most recently projecting $45 billion in savings to offset $12 billion in spending 2017–2026. Many have noted that even without these projections, $1 billion per year for CMMI is a drop in the bucket compared to the roughly $3 trillion spent per year on health care. Some of the specific models have shown financial promise as well, such as the risk-bearing ACO models and the hip/knee replacement outpatient payment bundle. Despite this, just two models—the aforementioned Pioneer ACO Model and the Medicare Diabetes Prevention Program (DPP) Model—have met the PPACA criteria for expansion.

As previously mentioned, the next battleground for CMMI will be regarding voluntary versus mandatory implementation of models. A new study from the University of Pennsylvania examined hospitals that participated in the voluntary BPCI model against those participating in the mandatory Comprehensive Joint Replacement (CJR) model. Researchers found that although the two programs engaged different types of hospitals (voluntary hospitals tended to be larger and higher-volume), there was no difference in baseline spending, quality of care or risk assumed. It is likely that hospital and physician groups will push back against any new mandatory programs. Whether Azar or Boehler will reinstate any remains to be seen.