Trendspotter: Pilot Sets Stage for New Payment Model

July 7, 2010

Medicare’s shared savings approach grew out of its Physician Group Practice (PGP) demonstration. The results of the pilot for first three years were fairly impressive.

The idea of accountable care organizations - provider-led entities that take various degrees of responsibility for the cost and quality of care - is starting to attract a lot of national attention. One reason for the interest is that CMS plans to start a shared-savings program involving ACOs in 2012. While the details are still vague, CMS will pay ACO participants fee for service plus a portion of shared savings above a certain threshold if physicians and other eligible providers meet its quality goals.

ACOs must include primary-care physicians and must coordinate care across all care settings. But they need not include a hospital. Organizations that might qualify as ACOs include hospitals with employed physicians, independent group practices, physician-hospital organizations, and clinically integrated IPAs. So there are opportunities for small practices to get involved in ACOs even if they’re not part of a larger organization. And, while the Medicare ACO program is still a couple of years off, some private health plans and self-insured employers are already starting to offer various incentive programs to ACOs.

Medicare’s shared savings approach grew out of its Physician Group Practice (PGP) demonstration. This pilot, which involved 10 large groups, began in 2005 and ended last March. Under the program, CMS paid the groups up to 80 percent of the shared Medicare savings from inpatient and outpatient care above 2 percent of historical costs. The additional payments were based half on efficiency and half on quality performance.

Among the participants in the PGP demonstration were:

• Billings Clinic, Billings, Montana Dartmouth-Hitchcock Clinic, Bedford, N.H.
• Dartmouth-Hitchock Clinic, Bedford, N.H.
• The Everett Clinic, Everett, Wash.
• Forsyth Medical Group, Winston-Salem, N.C.
• Geisinger Health System, Danville, Pa.
• Marshfield Clinic, Marshfield, Wisc.
• Middlesex Health System, Middletown, Ct.
• Park Nicollet Health Services, St. Louis Park, Minn.
• St. John’s Health System, Springfield, Mo.
• University of Michigan Faculty Group Practice, Ann Arbor, Mich.

The results of the pilot for first three years were fairly impressive. (Data is not yet available from 2009 and 2010.) By the end of the second year, all of the groups achieved benchmark performance on at least 25 of the 27 quality markers for patients with diabetes, congestive heart failure, and coronary artery disease. Four of the participants earned a total of $13.8 million as their share of the program savings of $17.4 million.

In the third year, all of the groups reached at least 28 of the 32 quality goals for the same three conditions plus hypertension and cancer screening. Five practices received a combined $25.3 million as their share of Medicare savings of $32.3 million.

Each of the participants approached quality improvement in a different way: The Billings Clinic, for instance, used registries and other electronic tools to ensure that evidence-based care was provided in every visit. Dartmouth-Hitchcock used the same methods and also hired experts to educate physicians and other staff members in the use of evidence-based guidelines. The Everett Clinic focused on improving hospital care, as well as transitions between inpatient and outpatient care and the use of imaging tests.

While these are all large, integrated groups, they still had to invest substantial sums in information technology and other infrastructure to achieve the desired quality and efficiency improvements. It wasn’t easy for them to recoup this investment, although the new infrastructure has helped some of them qualify for incentives offered by private payers.

The Geisinger Health System, for example, did not receive any shared savings in the first two years of the Medicare pilot but obtained $1.9 million in the third year. That was less than it spent on developing its capability to meet the quality goals, says Bruce Hamory, MD, chief medical officer emeritus of Geisinger. Nevertheless, Hamory believes that the demonstration was useful because it moved the organization in the right direction to take advantage of the emerging ACO and payment bundling trends.

It’s not yet known whether Medicare’s ACO shared-savings program will follow the same template that CMS used for the PGP demonstration.

But if it does, there are two lessons for physicians who are considering whether to participate in an ACO: First, bear in mind that you’ll have to meet quality goals to get any money. And second, because of the cost of the necessary infrastructure, it may take a while before the ACO recovers its investment.

But ACOs have one advantage that the PGP demo participants didn’t: the federal government will subsidize their acquisition of information technology to the extent that their physician and hospital members show meaningful use of electronic health records.