The federal accountable care organization (ACO) landscape continues to twist and turn, with one of the most recent developments being the drop of participants in the Medicare Shared Savings Program (MSSP), down from a membership high of 561 in 2018 to 477 at the start of 2021. Leaders at the National Association of ACOs (NAACOS) assert that the reason for the decline in program participants is related to several Trump administration-era policies, including 2018 changes from the Centers for Medicare & Medicaid Services (CMS) called “Pathways to Success,” which gave ACOs limited time before taking on financial risk and cut the share of savings most ACOs are eligible to keep. Last year, CMS also removed the opportunity for new ACOs to apply to join the Shared Savings Program in 2021, citing the pandemic. Nonetheless, NAACOS calls the MSSP the largest and most successful value-based payment program, pointing out that the model served 11.2 million seniors in 2019, and collectively saved Medicare $2.6 billion that year, and $1.2 billion after accounting for shared savings bonuses and collecting shared loss payments.
Moving up the federal ACO ladder, another program, the Next Generation ACO model, which includes Medicare’s most advanced, greatest risk-taking ACOs, is dealing with its own set of conflicting circumstances. Last month, NAACOS announced that 37 of the 41 ACOs in this model collectively saved Medicare $559 million in 2019, with four Next Gen ACOs left out because they’ve deferred financial settlement.
The program cares for 1.2 million seniors, but nonetheless, could be expiring soon. The Next Gen model was set to sunset at the end of 2020, though in light of the operational challenges brought on by the COVID-19 pandemic, CMS has extended the model’s final performance period through the end of December 2021. Another option for organizations that will now begin in April 2021 is the Direct Contracting model, which was originally intended to be the successor to the Next Gen program before CMS extended it. Direct Contracting creates three payment options for providers to take risks and earn rewards based on quality outcome, and leverages lessons learned from other Medicare ACO initiatives.
Despite its challenging structure, one organization that has had great success in the Next Gen model is Southwestern Health Resources Accountable Care Network (SWHR), the Dallas-based clinically integrated network which blends the strengths of University of Texas Southwestern Medical Center and Texas Health Resources. SWHR coordinates care for nearly 100,000 Medicare beneficiaries in North Texas aligned through the Next Gen ACO model, and saved more than $52 million in 2019 in the program. These generated savings place it at the top of organizations participating in the Next Gen model; SWHR has saved nearly $120 million since joining the program in 2017, its officials say. In a recent interview, Andrew Ziskind, M.D., senior executive officer of Southwestern Health Resources, discusses how his organization has been able to achieve success in a high-risk program, using data more effectively, and how the federal ACO landscape might play out in the future. Below are excerpts of that interview.
If could pinpoint a few core success factors for your achievements in this program, what would they be?
One of them is structural in that we have about 5,500 physicians in our clinically integrated network. So this is a physician-driven initiative, and we really strive to empower our physicians to partner with them, and in a sense wrap around them all of the infrastructure to do effective population health from a technology perspective, from a care management perspective, and from an analytics perspective. So strong physician ownership is one piece of it.
We're in a market that has relatively higher utilization rates than the rest of the country. Texas is number two after Louisiana in Medicare spend. So we focus very heavily on reducing unnecessary post-acute utilization. COVID has changed the world, but it takes a village with duct tape to connect all of the fragmented parts of the healthcare system, and we really try and take an end-to-end view of patient care to ensure that we're intervening early where we can reduce disease and reduce the need for admissions. And then once there is a need to contact the acute care health system, we work very hard to make sure that the handoffs are done well and that we're intensifying home services. Then we’re working very hard in areas that have the potential for overutilization, and some of the post-acute facilities in this market are very are known for high utilization.
ACOs rely on sophisticated data analytics to help meet quality and cost targets. Can you talk about the process involved in setting up that data infrastructure at SWHR?
For everybody across the country, using data more effectively is a work in progress. In general, the problem we have is not availability of data, but that people don't act on the available data. So now we are entering a period where we're moving from retrospective reporting to being able to effectively harness predictive analytics, and to some extent artificial intelligence. We need to be able to use our data to prospectively apply resources where they're needed in a timely fashion so there is maximum impact. If we can identify patients that are at higher risk of admission, readmission, or medication noncompliance, and start to care coordination around those patients, we will end up with better clinical outcomes and lower total costs of care.
We have had close to two dozen different systems feeding our care management processes. We’re in the process of finalizing the implementation of Arcadia as a population health platform. We really believe that provider networks have a huge advantage because now we have the ability to integrate clinical data and claims data, and present that information at the point of care in a way that identifies gaps, triggers, reminders, and has analytics driving alerts. We think that that is really the next generation of using data to make the work on the physicians easier.
The critical element is how we can identify patients who need additional resources and support, whether it's home-based interventions, pharmacy support, etc.—at any level. [Often], you wait until someone is already hitting the healthcare system. They're in an emergency department, they’re admitted, they've had surgery, they've had some acute process going on, and when we identify those folks, more and more we've been relying on an inpatient care coordinator to start that process as soon as the patient is in an inpatient facility. That’s when we can then start to apply resources. But [ideally], we would much rather prevent the accidents or prevent the need for some of this.
We're starting to integrate data from multiple sources. We're focused a lot on social determinants of health (SDOH), and I feel very strongly that although many people think of social determinants of health as economically disadvantaged populations—and that's absolutely true—SDOH occur across all socioeconomic strata, across all racial groups, and across all different cultural backgrounds. And we need to be sophisticated enough to contour our care management resources around the social needs of a patient and meet the patient where they are.
CMS had planned to sunset the Next Gen program at the end of last year, but extended that decision due to the pandemic. Have you heard anything regarding the program’s future?
Right now, our understanding is that the Next Gen program will be sunsetted at the end of 2021. There are three potential buckets. There's the MSSP, Direct Contracting, and the Geographic Direct Contracting model. We’re actively exploring what would be our successor plan to Next Gen, but at the same time we also have a change in administration. So there's the potential for an extension or a change in the proposed requirements for these programs. There are a lot of unknowns right now. We are coming up on application deadlines, and everybody's running in circles trying to figure out what to do.
If we step back a moment, the shared savings programs [in Medicare] are transitional and meant to change a market. You can't achieve shared savings in perpetuity; it's virtually impossible to keep lowering the cost of care to zero, taking an extreme view. So they have to be transitional and the question is what are the long-term sustainable models? Today, what's striking is that we are not seeing a decrement in savings, but we're continuing to see a growth and savings, which we will take credit for. It means we have implemented the kind of population-based resources that really do achieve savings. And people could say we're in an advantaged market because there is more overutilization in the Dallas/ Fort Worth market, but at the same time I would say we are in a highly deregulated market with a lot of medical entrepreneurship and a culture of autonomy [among] the physicians that makes driving change harder. So on the one hand we have more cream at the top and on the other hand, we have a very difficult set of clinical care patterns to change.
What advice can you give others who have yet to or are unsure of entering into these types of programs?
One would be to avoid magical thinking. There are a lot of organizations that believe you can print the banner on the door, that you can sign up the doctors, and everything will work. This is hard work and requires a thoughtful, integrated approach that addresses several things: care coordination, growing pharmacy costs, the social determinants of health, using data more effectively and analytics, physician engagement, and change management. And, in the absence of a robust comprehensive program, you're not going to get the kind of results.
There are health systems that historically have dominated their markets, and are used to success in virtually every initiative they encounter. Then they take risk and find that it really does require a lot of infrastructure. So, [we are] focusing on reducing total cost of care and improving outcomes, and in total cost of care we've demonstrated nearly a 20 percent reduction in outpatient visits, unnecessary outpatient visits, and more than 10 percent lower ER visits.
But when I think about effectiveness of a clinically integrated network that can contract us an ACO, I think we are moving toward a new generation of relationships with physicians. And we have to move from business relationships to engaging the hearts and minds of physicians in a way that improves their ability to practice medicine and that improves their professional satisfaction. So I'm talking new models of physician engagement and support, not just financial relationships. We really have to be part of transforming healthcare to fundamentally make it better.