Should we do away with ACOs? The answer depends on who you ask  

Though accountable care organizations have been around for more than a decade, their effectiveness at reducing healthcare costs remains a highly debated issue, prompting impassioned condemnations and defenses from experts in the industry. A new study has stirred up the pot once again.

The goal of ACOs, that is, entities that take responsibility for the quality and cost of care across settings, is to move the healthcare system toward value-based care. The ACO model aligns incentives with a focus on reducing costs and improving care so participating providers can share the savings earned. The Centers for Medicare & Medicaid Services launched its first ACO payment model in 2012.

But a new study published in the Journal of General Internal Medicine suggests that ACOs may not be cutting costs as expected. CMS’ ACO programs roughly broke even — which means, the programs either lost money or saved no more than a few tenths of a percent.

For the study, researchers compared financial performance data from all four CMS ACO programs from 2005 to 2018.

“We were impressed by the consistency of the poor ACO effects across all program variations,” said Dr. James Kahn, a study author and professor at the Philip R. Lee Institute for Health Policy Studies at the University of California San Francisco, in a phone interview. “This is not a case of ‘this version didn’t work but this did’ — this is ‘ACOs failed to make much of a difference each time.’”

Kahn and his co-study author Kip Sullivan, a member of the advisory board of community organization Health Care for All Minnesota, are calling for the end of the ACO model.

“ACOs have not worked for Medicare and will not work for the broader healthcare system,” they wrote in the study.

They doubled down on this idea in an opinion piece published on STAT News last week. Instead, they suggest relying on lessons learned from other countries that have adopted single-payer systems and universal coverage.

But others in the industry pushed back on this notion, starting with the National Association of ACOs.

“ACOs aren’t going anywhere,” said David Pittman, senior policy advisor at the association, in an email. “ACOs enjoy bipartisan support, and payment reform is needed to move us away from a system that financially rewards volume-based care. ACOs and value-based care aren’t going anywhere and, in fact, are needed now more than ever.”

Pittman took issue with both the study’s conclusion and its methods.

The research excludes the most recent data — the study stops at 2018 —which are ACOs’ most successful years, he said.

Just last week, CMS announced that ACOs participating in the Medicare Shared Savings Program in 2020 earned savings totaling nearly $2.3 billion while saving Medicare approximately $1.9 billion. This is the fourth consecutive year of net savings for Medicare, CMS said in a news release.

When asked why they did not include data beyond what was available in 2018, Kahn pointed to an article by professors at Harvard Medical School and the University of Southern California.

The article shows that “2019 net savings are minimal despite selection and attrition bias,” Kahn said. “We could have included a bit more time, but it wouldn’t have changed the findings.”

Selection bias refers to ACOs that only enter CMS programs if they are confident in their ability to generate savings sooner, whereas attrition bias refers to a greater drop-out of ACOs that failed to achieve savings.

Dr. J. Michael McWilliams, the Warren Alpert Foundation Professor of Health Care Policy at Harvard Medical School who co-wrote the article that Kahn referenced, disagreed with his conclusion that the ACO model should be dismantled.

Though it is true that savings have been modest and the incentives to save have been weak, the answer is not to end the ACO model altogether.

“The more productive policy strategy is to improve the design of the models to build on early progress, as opposed to jettisoning them based on superficial considerations of their success to date,” said McWilliams in an email.

The second part of his article jointly written with Dr. Alice Chen, an associate professor of public policy at the USC Price School of Public Policy, delves into strategies for improving the ACO models implemented by CMS.

Like NAACOS’ Pittman, McWilliams finds Kahn and Sullivan’s study to be lacking.

“The study in [Journal of General Internal Medicine] is not a rigorous one,” he said. “It is an opinion piece dressed up as research.”

Dr. John Hsu, director of the clinical economics and policy analysis program at Massachusetts General Hospital’s Mongan Institute for Health Policy, echoed McWilliams’ criticisms.

“The impact of the ACO program on spending depends heavily on the nature and magnitude of the incentives applied,” said Hsu in an email. “To date, these incentives have been mild and generous in part because provider organizations drop out of the voluntary ACO program if they started to lose money or believed that the program incentives created an onerous system.”

Though Hsu believes it is valuable to have a summary of available estimates of the ACO impact, which is what Kahn and Sullivan’s study attempts to do, he challenges their conclusions about what these estimates mean for the ACO program.

“The ACO program sought to alter care delivery by changing the incentive system,” he said. “This type of policy change by definition takes time to have an impact because it is inherently indirect.”

It is important to note that Hsu’s employer —Mass General Brigham, Massachusetts General Hospital’s parent — operates two ACOs. While the Mass General Brigham Medicare ACO generated savings in 2019, the first year that it participated in the Medicare Shared Savings Program, the health system spokesperson was not able to provide data showing the second ACO, called the Mass General Brigham MassHealth ACO, also saved money.

Though there is no conclusive evidence as to the efficacy of ACOs, CMS is continuing to experiment with the model. In May, CMS announced that its Next Generation ACO Model will end Dec. 31. The model enables participants to assume higher levels of financial risk and reward than are available under the Medicare Shared Savings Program.

Instead, the agency will allow Next Generation model participants to apply for the Global and Professional Direct Contracting Model, a set of two voluntary risk-sharing options that, like the ACO model, aim to reduce expenditures and enhance the quality of care.

The future of ACOs aside, it’s safe to say that tinkering with different models to attain the overarching goal of improving care and reduced costs, will continue.

Photo: Dutko, Getty Images