AMA Journal of Ethics^®

Iglehart JK. Assessing an ACO prototype: Medicare's Physician Group Practice demonstration. N Engl J Med. 2011;364(3):198-200.

Wilensky GR. Lessons from the Physician Group Practice Demonstration: a sobering reflection. N Engl J Med. 2011;365(18):1659-1661.

In August 2011 the Centers for Medicare and Medicaid Services (CMS) released findings from its Physician Group Practice (PGP) Demonstration, “a landmark partnership with physician group practices that aims to better coordinate care across different settings, leading to improved quality and cost savings” [1]. At the end of the 5-year project, CMS announced that the demonstration showed such “significant progress in areas of both quality improvement and savings in Medicare expenditures” that it has extended the original program into a 2-year supplementary transition demonstration [1].

Medicare’s first pay-for-performance initiative launched in 2005 with 10 PGPs and was designed to help coordinate Part A and Part B Medicare services, “promote cost efficiency and effectiveness through investment in care management programs,” and “reward physicians for improving health outcomes” [2]. The key point of the demonstration is that physicians still received their Medicare payments but also had the opportunity to receive additional performance payments if they generated savings and satisfied benchmarks on 32 quality measures. According to CMS Administrator Donald M. Berwick, the PGP demonstration “provide[s] great insight into how to use Medicare’s payment systems to improve quality while reducing costs” [1]. There is some disagreement, however, over the promise of the PGP demonstration to lead to greater savings and to what degree it should serve as a model for accountable care organizations (ACOs).

In her recent article, “Lessons from the Physician Group Practice Demonstration—A Sobering Reflection,” Gail Wilensky argues that the overall results of the PGP Demonstration project are cause for some concern [3]. After providing a brief overview of the demonstration and discussing some of the (limited) benefits of the project, Wilensky questions whether it generates enough savings to serve as a viable cost-cutting model for ACOs going forward. The “good news” about the project is that it did “very well” in meeting or exceeding nearly all of the 32 quality goals, which assessed performance and quality measures in areas like preventive care, diabetes mellitus, coronary artery disease, and hypertension [3]. But while these results might be “worth applauding,” Wilensky believes they are overshadowed by the fairly unremarkable savings generated by the PGP participants.

For example, after the first year of the demonstration only two PGPs managed to exceed the 2 percent savings benchmark, and only half of them did so after 3 years [3]. According to Wilensky, such meager savings do not bode well for ACOs, which will also have a 2 percent saving threshold but “will have to meet benchmark levels on an even larger set of quality metrics than the PGPs did” [3]. Essentially, Wilensky’s concern is that, if the 10 practice groups selected for the demonstration had such difficulties improving the quality of health care services while also improving cost efficiency (e.g., exceeding the 2 percent savings threshold), how could this serve as an effective model, on a much larger scale, for ACOs? Wilensky remains skeptical that ACOs, as they are currently envisioned by the CMS, will be a “viable alternative to both traditional Medicare and traditional managed care” that can “encourag[e] newly formed groups to provide care in ways that can both improve quality and reduce costs” [4].

While Wilensky takes the pessimistic glass-half-empty approach to modeling ACOs on the PGP demonstration, others are more optimistic. Donald Berwick, for instance, believes that the ACO model will benefit from the PGP demonstration because it:

helped to identify several factors that are critical to improving quality and increasing the opportunities for shared savings: an integrated organization that supports expending resources on programs to improve quality and reduce the provision of unnecessary services; dedicated physician leadership with a proven ability to motivate the implementation of quality-improvement programs; and a central role for health information technology in enabling the organization to manage population health and receive feedback at the point of care [5].

But while such lessons can be gleaned from the PGP demonstration and applied to the long-term implementation of an ACO model, the key issue is identifying what else can be done in the meantime to help curb Medicare spending. CMS is currently scheduled to launch the ACO program in January 2012, but it is unclear how long it will take for ACOs to show that they are successful in reducing Medicare spending. As John Iglehart opines in his article, “Assessing an ACO Prototype—Medicare’s Physician Group Practice Demonstration,” the “ACO model is a work in progress, and the CMS must address many questions in crafting the new program’s regulations” [6].

Overall, Iglehart views the demonstration as a successful model for reforming Medicare—based on the limited data of the 10 PGPs, the project showed promising results in improving the quality of care while generating cost savings. Iglehart’s worry, however, is that Congress will need to act to implement some type of stopgap measure to help increase Medicare savings before it is clear whether or not ACOs can be as effective as the PGP demonstration in improving care and reducing overall costs [6]. It is simply not enough to have faith that ACOs will single-handedly solve Medicare’s ongoing financial crisis.

The current debate about the future of Medicare only promises to turn more contentious as discussions about the sustainable growth rate become more urgent, as CMS announces its final rule on ACOs, and as health systems across the country begin to spend more time and money preparing to transition to the ACO model by January 2012. As Donald Berwick comments, “accountable care is not panacea” for reforming Medicare but is “one of a number of complementary initiatives chartered by the ACA [Accountable Care Act] to help achieve the three-part goal of lower costs, improved care, and better health” [5]. What remains to be seen, however, is just how these “complementary initiatives” can work in concert to help make Medicare an effective, efficient, and sustainable program. It is up to the government and health care systems throughout the country to make it happen.