All Content

Consider ACO Participation As Medicare Weighs Changes to Shared Savings Program


In December 2014, nearly three years since its launch, the Centers for Medicare and Medicaid Services (CMS) issued the first proposed rule changes to the Shared Savings Program. The changes, if approved, would take effect in the 2016 performance year and would focus on a host of alterations impacting participating accountable care organizations (ACOs), including reduced administrative burden, improved function and transparency, and enhanced incentives to participate in risk-based models.

Experts say the changes could address some of the biggest flaws in the program but also may not go far enough to incentivize more healthcare providers to participate—or protect them from the risk of financial loss. The rules are under review following a public comment period.

“Many features about the original rules weaken the incentives to participate in ACOs,” says Michael McWilliams, MD, PhD, associate professor of healthcare policy and medicine at Harvard Medical School and a practicing primary care physician at Brigham and Women’s Hospital in Boston.

The ACO model encourages providers to realize savings under fee-for-service Medicare through better-coordinated care and improvements in metrics related to utilization and quality. Any savings relative to a benchmark year are shared between the ACO and CMS.

This year, 424 ACOs are participating in the program nationwide, and while the number of Pioneer ACOs has fallen in recent years (Pioneer ACOs wager higher savings for participants at the risk of greater financial loss), a new independent report commissioned by CMS shows the program saved more than $300 annually per beneficiary in its first two years, achieving $384 million in savings.1 In a statement, CMS concluded this meets the criteria for expanding the Pioneer program; however, Dr. McWilliams says policy changes may still be needed to encourage participation in ACOs with downside risk.

In January, Dr. McWilliams and colleagues published a study in Health Affairs that demonstrated that existing benchmark rules may actually encourage higher Medicare spending as ACOs try to “fatten up” so they have more improvements to make and, therefore, more chance of success at realizing savings.2

Currently, providers’ performance is stacked against their performance and cost benchmarks established in the year prior to forming an ACO. As improvements are made, it becomes increasingly challenging for ACOs to do better. Dr. McWilliams says ACOs should instead be compared to other ACOs and providers.

It’s a “melting ice cube problem,” says Gregory Burke, MPA, director of innovation strategies for New York-based United Health Fund (UHF), a research and philanthropic organization focused on advancing healthcare.

For a hospital system, the bulk of the money comes from inpatient care. We’re saying: ‘You stomp on your own air hose today, and a year from now I’ll give you 50% oxygen—and you have to share with your buddy.’

—Gregory Burke, MPA, director of innovation Strategies, United Health Fund

“You are punishing the good, lean providers that are efficient,” he adds, “and rewarding people who are less efficient, in terms of cost of care and utilization of services.”

Burke and a colleague at UHF, health policy analyst Suzanne Brundage, recently completed qualitative and quantitative reports on ACOs in the state of New York, which currently make up 20% of Medicare fee-for-service beneficiaries.3,4

Through their analysis, which included structured interviews with 17 Pioneer ACO leaders, Burke and Brundage found ACO rules could change in the following ways to make the program sustainable and more attractive to providers:

  • Patients should be attributed to PCPs within the ACO;
  • Risk adjustment should be made for ACO providers serving a sicker population of patients; and
  • Benchmark rules should be altered.

Additionally, Dr. McWilliams says the shared savings rate realized by ACOs should be higher than 50%, which is especially true for hospitals within an ACO, since the goals of the program are to reduce hospital visits, extensive specialist services, and testing services.

“For a hospital system, the bulk of the money comes from inpatient care,” Burke says. “We’re saying: ‘You stomp on your own air hose today, and a year from now I’ll give you 50% oxygen—and you have to share with your buddy.’”

The 2014 State of Hospital Medicine report indicates that 36% of adult hospitalist medicine groups are in hospitals either already involved in or considering involvement in an ACO; however, respondents in that report also reflect no clear role for hospitalists in the ACO model.

This is a point disputed by Val Akopov, MD, vice president and chief of hospital medicine at WellStar Health System, a not-for-profit organization in northwestern Atlanta and a participating ACO. Dr. Akopov highlights ways in which hospitals and hospitalists could take advantage of the model.

“Five measures fall into the domain of care coordination that are directly, unequivocally related to what hospitalists do, and these metrics are part of, in my opinion, what any hospital medicine program should have as a value proposition,” Dr. Akopov says.

At WellStar, for example, hospitalists have become part of the ACO structure by serving as medical directors and attending physicians at skilled nursing facilities (SNFs). They are “solely responsible to attend to patients in SNFs, and we have seen a dramatic improvement in readmission rates and quality metrics in nursing homes, such as incidence of falls, use of antipsychotics, and 30-day unplanned readmissions to acute care hospitals,” Dr. Akopov explains.

Additionally, WellStar hospitalists work with each inpatient to ensure they have primary care follow-up scheduled before discharge, Dr. Akopov says, noting that the model is a good opportunity to explore changes to the way hospitals and providers deliver care.

“There are roughly 38,000 Medicare patients in [our] ACO; it’s much easier to work out the kinks with innovations on a limited patient population and then extrapolate findings on 1.5 million annually, rather than trying to bite too much,” he says.

Despite the challenges, experts are optimistic the ACO model can—and will—work. In their reporting, Burke and Brundage found healthcare leaders participating in ACOs remain optimistic.

“It’s a post-Copernican universe, where the world no longer revolves around the hospital, so balancing the equation is a little different,” Burke says. “But they’re staying in the game, because that’s where the puck is going to be.”

Kelly April Tyrrell is a freelance writer in Madison, Wis.


  1. Affordable Care Act payment model saves more than $384 million in two years, meets criteria for first-ever expansion. Centers for Medicare & Medicaid Services website. Published May 4, 2015. Accessed May 11, 2015.
  2. Douven R, McGuire TG, McWilliams JM. Avoiding unintended incentives in ACO payment models. Health Aff. 2015;34(1):143-149.
  3. Burke, G, Brundage S. Accountable care in New York state: emerging themes and issues. United Hospital Fund. Accessed May 9, 2015.
  4. Burke, G and Brundage S. New York’s Medicare ACOs: participants and performance. United Hospital Fund. Accessed May 9, 2015.

Recommended Reading

Hospitalist Consults on Psychiatric Patients Concern Nurses
The Hospitalist
Hospitalists Should Make Commitment to Improve Healthcare Safety
The Hospitalist
Physician Culture, Beliefs Drive Variation in Healthcare Spending
The Hospitalist
Many Physicians Unaware of Costs for Lab Tests, Procedures
The Hospitalist
Insufficient Time for Patients May Add to Healthcare Waste
The Hospitalist
Hospitals’ Uncompensated Costs Estimated at $27.3 Billion in 2014
The Hospitalist
Hospitals with Hotel-Like Amenities Don’t Improve Satisfaction Scores
The Hospitalist
Quality Data Dashboards Provide Performance Feedback to Physicians
The Hospitalist
Why Physicians Override Best Practice Alerts
The Hospitalist
How Should a Patient with Pulmonary Hypertension Be Evaluated, Managed?
The Hospitalist
   Comments ()