Testimony on Alternative Payment Models and the Slowdown in Federal Health Care Spending

Chairman Whitehouse, Ranking Member Grassley, and Members of the Committee, I appreciate the opportunity to appear before you today. In consultation with Budget Committee staff, I have focused this testimony on accountable care organizations (ACOs), the Center for Medicare & Medicaid Innovation (CMMI), and the unexpected slowdown in federal health care spending.

Why Might ACOs or CMMI Have Contributed to Reducing Federal Health Care Spending?

Because ACOs voluntarily assume responsibility for the quality and costs of care for a defined group of patients, they have the potential to reduce unnecessary care, improve care coordination and patients’ health, and reduce spending. Health care providers participating in ACOs or other value-based payment arrangements receive financial incentives to improve the efficiency and quality of care. Such incentives contrast with those found in Medicare’s fee-for-service program, in which separate payments are generally made for each encounter or service delivered.1 Reimbursing on a fee-for-service basis tends to create incentives for providers to deliver additional and more-complex services, potentially contributing to the high costs and uneven quality of health care.

CMMI’s goal is to identify approaches that reduce spending or improve the quality of care. To do so, it operates models that test new ways to deliver and pay for health care, including models that establish value-based payment arrangements with providers.

What Are the Effects of the MSSP on Medicare Spending?

The Affordable Care Act (ACA) established a permanent ACO program in Medicare known as the Medicare Shared Savings Program (MSSP). The Congressional Budget Office reviewed evidence on the performance of the MSSP and found that the program was associated with small net budgetary savings in the early years of its operation.2 To arrive at that assessment, CBO largely relied on peer-reviewed literature that compared changes in spending among beneficiaries attributed to ACOs with changes in spending among a control group of beneficiaries not attributed to an ACO. More recent evidence is limited and is challenging to interpret. It has become increasingly difficult to find a reasonable control group to use in evaluations, and evaluations are complicated by providers’ ability to opt in and out of the program.

What Were CMMI’s Budgetary Effects Over Its First Decade of Operation?

CBO has updated its estimate of the budgetary effect of CMMI and currently estimates that CMMI’s activities increased direct spending by $5.4 billion, or 0.1 percent of net spending on Medicare, between 2011 and 2020.3 Specifically, CMMI spent $7.9 billion to operate models, and those models reduced spending on health care benefits by $2.6 billion. The estimate reflects CBO’s review of published evaluations of 49 models initiated in CMMI’s first decade after it was established under the ACA, as well as corresponding historical budget data.

What Are CBO’s Projections of CMMI’s Effects Over the Current Baseline Projection Period?

CBO estimates that over the current baseline projection period, 2024 to 2033, CMMI will increase net federal spending by less than $50 million. Over that period, the estimated effect of CMMI’s activities transitions from an annual net increase to an annual net decrease, reflecting ongoing growth in the number of certified models that continue to produce savings over time.

How Does CBO Estimate the Effects of Legislative Proposals to Change CMMI?

Legislative proposals that the Congress could consider that would affect CMMI fall into one of three categories: modifications to specific models, changes to the parameters within which CMMI operates, and a repeal of CMMI’s statutory authority or rescissions of unobligated funding. In general, CBO’s analysis considers available evidence on specific models. When such data are not available or the legislation is not related to a specific model, CBO relies on a more general framework using information about CMMI’s prior activities and performance. Estimated effects would depend on the details of the legislation.

Have Alternative Payment Models Contributed to the Slowdown in Health Care Spending?

The implementation of the MSSP and the creation of CMMI have occurred during a period of unexpectedly slow growth in federal health care spending. Whether that slow growth is related to CMMI or other alternative payment models created under the ACA is not entirely understood, but CBO’s review of the effects of the MSSP and its estimate of the budgetary effects of CMMI’s first decade of operation suggest that they were not factors. Still, some researchers have posited that the existence of CMMI may have led to broader systemwide changes that are not attributable to a specific model. Such changes may have led to increases or decreases in federal health care spending, which are not reflected in CBO’s estimates of the budgetary effects of CMMI.4

Over the past decade, CBO has been tracking the slowdown in federal health care spending and has previously pointed to several contributing factors.5 Broader factors include decreases in the growth of Medicare’s payment rates, reduced spending on patients with cardiovascular diseases (because of better management of those conditions and greater use of medications to control risk factors), and a shift in the relative importance of technology in fueling the growth of health care spending. Federal spending on the Medicare and Medicaid programs also grew more slowly than CBO projected. A key factor underlying Medicare’s slower-than-expected growth was slower growth in net spending on prescription drugs; for Medicaid, a key factor was less-than-anticipated spending for long-term services and supports.