Correcting the course of value-based care models

Notwithstanding years of federal efforts, the rate of healthcare cost growth has continued to accelerate, putting care out of reach for too many Americans. Employers and state Medicaid programs struggle to keep up, and parts of the Medicare program are projected to run short of funding.

Value-based care is widely considered the solution. It moves away from fee-for-service and instead aligns financial incentives to produce higher quality and better results. The Trump Administration has recognized, however, that important policy adjustments are necessary if this highly anticipated transition is finally to take place.

From the beginning, we have taken actions to promote value-based care across all our programs. CMS has identified value as an agency-wide strategic objective, and efforts are underway across the agency to promote it.

That includes our historic efforts on price and quality transparency, which empower patients by creating a market in which providers must compete for them on the basis of cost and quality. Similarly, our push for interoperability of electronic medical records advances innovation, drives seamless, coordinated care, and promotes evidence-based treatments—hallmarks of value-based care.

CMS’ proposed regulations to modernize the Stark Law also do much more to encourage value-based arrangements between providers and promote more care coordination.

Importantly, we are integrating Medicaid—left behind for too long—into the transition to value. In a letter to every state Medicaid director in the country, we recently offered a range of innovative ideas to introduce more value-based incentives into state programs. Medicare can’t be on this journey alone; we need every player in the healthcare system, public and private, to be engaged.

But the transition to value also demands changes to the Center for Medicare and Medicaid Innovation (CMMI), established by Congress a decade ago specifically to promote value through the testing of alternative payment models. CMMI’s reach is considerable: it has developed 54 payment models—many of which the Trump Administration inherited—and more than 450,000 providers participate, serving over 26 million people.

Nevertheless, an appraisal of CMMI’s record is deeply concerning: only five models have shown statistically significant savings, and only three have met the criteria for national expansion. Evaluations show they have lost billions of dollars, and just a handful have seen significant improvements on quality metrics—a weak return on investment for taxpayers. We have been studying what has worked and what is contributing to these losses in order to turn CMMI around.

First, it’s clear that voluntary models designed with an abundance of financial carrots to attract participation are often insufficient to avoid significant losses. Upfront additional payments have also stymied models’ financial success. After accounting for the enhanced payments to providers, CMMI has shown net losses. Models where providers have downside risk perform better because they have “skin in the game.”

That said, more regulatory flexibility is one carrot that can work. We believe many of the waivers provided during the pandemic could be offered in models to attract participation.

Second, many CMMI models allow participants that reduce spending relative to a target or benchmark, to receive a share of the savings to Medicare from CMS. But this payment structure only works for Medicare if the targets are accurate. Because that has not always been the case, CMS has paid out too much to model participants, resulting in net losses despite a significant gross savings.

We saw this dynamic at work in multiple models including the Comprehensive Care for Joint Replacement model, the Next Generation ACO model, and others.

We are already successfully implementing these lessons learned and adjusting models that have problems. In 2018, the Pathways to Success final rule restructured the participation options under the Medicare Shared Savings Program to require groups of healthcare providers that form an accountable care organization to take on risk to continue participating, while providing more flexibilities to those that do. Since then, the Shared Savings Program has showed continued Medicare savings, including $1.19 billion in 2019, the highest ever in a single year.

Last year, we announced the CMS Primary Cares Initiative featuring the Direct Contracting model, which has rigorous, data-tested benchmarks and a flexible design that encourages providers to take on more risk.

Just last month, after discovering the model was resulting in significant net losses, CMS adjusted the Bundled Payments for Care Improvement Advanced Model for the Model Year beginning Jan. 1, 2021. Going forward, we intend to implement more mandatory bundled payment programs that build on the best components of the BCPI model while addressing selection effects and other factors.

Finally, CMMI must provide participants with timely access to data and analytics. Sharing data with participants allows for earlier, data-informed interventions. This approach worked well in our Data at the Point of Care pilot program, which connected claims data directly to providers in their EHRs.

Taken together, these changes can help CMMI correct course and fulfill its considerable promise. The coming adjustments will at times upset the status quo, but we should remind ourselves Americans deserve a healthcare system that rewards providers for keeping them healthy.

CMS will continue to push value not just in CMMI but across all our programs because patients are better served by a healthcare system that rewards for outcomes, not just volume. We are optimistic that over the next decade the promise of value-based care can be realized.