What Research Says About ACO Success
Taking on downside risk can be a big motivator for organizations to take more seriously the transformation needed to succeed in value-based care, say researchers who study accountable care organizations.
Speaking during a recent webinar put on by the Primary Care Collaborative, Jonathan Gonzalez-Smith, a research associate at the Duke Margolis Center for Health Policy, said, “ACOs in two-sided risk models are more likely to invest in care management services, develop better data management capabilities, or establish care coordination and the staff required to manage their beneficiaries’ care.”
Gonzalez-Smith helps lead the Margolis Center’s research on international models of accountable care, health financing, global health innovation, and payment and delivery reform. His research evaluates how to support health system transformation at the organizational, regional, national, and international level to achieve better population health while promoting efficiency, equity, and high-quality care.
He was joined by David Muhlestein, Ph.D., J.D., chief research officer for Leavitt Partners, where he directs the study of accountable care organizations through the LP Center for Accountable Care Intelligence and leads the firm's quantitative evaluation of healthcare markets. The two were interviewed by Mai Pham, M.D., M.P.H., who previously served as chief innovation officer at the Center for Medicare and Medicaid Services.
Gonzalez-Smith noted that a lot of the research and discussion to date has focused on the Medicare Shared Savings Program, in part because the vast majority of evidence comes from MSSP. It is studied heavily because the data is publicly available, and it's also a foundational program that has helped drive ACO participation across other payers. “It is important to note that while several population-based payment models, including MSSP, have contributed to increases in primary care participation, they have not caused significant movement of primary care as a whole into value-based payment arrangements. Part of that is tied to the challenges related to the resources, time, and know-how required, which create impediments to entry for many smaller primary care-based practices.”
Approximately 40 percent of total U.S. healthcare payments were tied to some sort of advanced value-based payment model in 2020. This varies by payer, with higher rates of adoption in Medicare than in Medicaid, Gonzalez-Smith said. “Our research has found this is beginning to change. Commercial and Medicaid contracts have continued to increase steadily, and part of this has also been driven by flexibility in the commercial market and the increasing number of states that have implemented managed Medicaid managed care programs, with at least 40 states using some form of managed care for their Medicaid population as of 2019.”
ACO growth, particularly in the Medicare program, has plateaued in the last few years. This can be attributed to several factors, he said. There's the overall focus from CMS and CMMI to encourage organizations to move into more downside risk arrangements. This was demonstrated in the 2019 Pathways to Success rule, which overhauled MSSP and truncated the time that organizations were required to move into downside risk arrangements. This increase in downside risk adoption, particularly in the voluntary models, has kept organizations from participating.
Programs such as Direct Contracting also may have dampened enthusiasm as some providers stayed on the sidelines waiting to see where the chips are going to fall, Gonzalez-Smith said, adding that the pandemic was a significant disruption for provider organizations that were thinking about moving into value-based payment arrangements. “We've seen that those organizations already operating in value-based care arrangements are much more financially resilient and better prepared to respond to the pandemic than those in fee-for-service arrangements,” he added. “We've also seen that the stagnation was a trend that preceded the pandemic. One of the reasons is that it takes time to be successful. We found that ACOs have an initial learning curve and need a period of about one to three years to get acclimated to the MSSP program.”
ACOs that are more experienced seem more likely to achieve savings, with about 52 percent achieving savings in a third year and 83 percent doing so in a fourth year of participation, according to Gonzalez-Smith. Part of this might be attributed to survivor bias as those that aren't doing well drop out after their contracts terminate, he noted. But it also takes time to develop the capabilities and the technical infrastructure to forecast utilization trends and predict benchmarks. “We've seen both from a qualitative and quantitative perspective that it is a challenge for smaller physician-led and primary care practices with limited resources to redesign clinical processes and invest in the infrastructure and the data analytics. Some have hired actuarial firms to help forecasts costs, though this can be expensive. Others have worked with third-party firms that we're calling ACO enablers like Aledade, or Caravan, or others.”
There has been an increase in the adoption of two-sided risk models. Thirty-seven percent of MSSP ACOs were in two-sided risk in 2020. This has been driven in part by the inherent design of MSSP and the stated goal of CMS to shift more organizations into downside risk, Gonzalez-Smith noted. But this has had a material impact on performance and savings, with ACOs in two-sided risk models more likely than those in one-sided risk models to receive bonuses and generate savings. Part of this, again, may be attributed to survivors’ bias with organizations that are already performing well tending to stay in the program.
The Margolis Center’s research has also shown that slightly more ACOs in upside-only arrangements did indeed drop out of the program than those in downside-risk arrangements, which is a trend that's held up over time. “But it's also because ACOs that are able to invest in developing their capabilities have improved and are making the changes needed to succeed in MSSP,” Gonzalez-Smith said. Downside risk can be a big motivator for organizations to take more seriously the transformation needed to succeed. Downside risk can be an effective catalyst to help drive care delivery transformation.”
Looking ahead, there are a number of developments at the federal level that will have significant impact for the direction of ACOs, he added. CMS has explicitly stated that it wants to move not just Medicare but Medicaid beneficiaries into a care relationship with accountability for quality and total cost of care by 2030.
Gonzalez-Smith said there will be more attention to mandatory population-based payment models, which will be a linchpin of the strategy, “because it can help mitigate risk selection as well as promote care coordination,” he said.
He re-emphasized that “downside risk is a big motivator for undertaking big care transformation changes,” and added that there will need to be better multiplayer coordination. “Having just one payer on board isn't sufficient, and our research has shown that organizations that do better often have more than one contract in some kind of value-based payment contract.”
Gonzalez-Smith added that it is good to keep in mind that payment reform is not an end goal in itself, and that further work is needed to continue to engage patients and ensure that they're the focus of care delivery and payment reforms, and improving access and outcomes while lowering costs.
Muhlestein has studied some of the reasons that ACOs are not successful, and he said there are lots of reasons some ACOs are not able to save money and quit the program. “The fastest way to be unsuccessful is to invest significant resources into initiatives that don't actually help,” he said. “If you spend a lot of money for patients who don't need it, then you're going to cost more and you're not going to be able to save money. Another challenge is that sometimes the operational side. Perhaps the leadership initially says yes, let’s move forward with an ACO, but then when they get into the details, they realize there are too many organizational barriers that keep them from providing that optimized care that you'd like to see happen under one of these models. Also, many organizations see transitions in senior leadership and in their boards, and this priority of focusing on value-based care is not always consistent between one leadership group to the next. “We're seeing people who say, ‘well, it was a great experience or experiment, but we're going to focus on something else.’”
Gonzalez-Smith and Muhlestein said both physician-led and hospital-led ACOs can succeed in value-based care. “The reality is that both hospital-led and physician-led ACOs do quite well. On average, physician groups do slightly better. If I can simplify everything that we've seen about ACOs in a nutshell, it is that more primary care leads to better performance, as long as it's able to successfully reduce hospital admissions and skilled nursing admissions. That's really what the key is.”
