Significant Potential for Accountable Care Organizations to Improve Care, Lower Costs

Media Contacts:
Annmarie Christensen
(603) 653-0897

JAMA study analyzing quality, spending, and savings across pilot shows promise in new reforms

Lebanon, N.H. (September 11, 2012) – New health care delivery models that reward providers for coordinating and improving care hold promise for slowing the cost of treating the sickest, costliest patients in the health care system, according to a new study by Dartmouth researchers published in the Sept. 12 issue of the Journal of the American Medical Association.

To learn how new delivery models, such as accountable care organizations (ACOs), are likely to perform for patients with severe health conditions, researchers from the Dartmouth Atlas Project and The Dartmouth Institute for Health Policy & Clinical Practice studied a similar payment reform model—Medicare’s Physician Group Practice Demonstration (PGPD). The study focused on the care provided to patients covered by both Medicare and Medicaid, referred to as “dual eligible” patients.

The nation’s 9 million dual eligibles comprise 20 percent of the Medicare population but account for 31 percent of its spending, and comprise 15 percent of the Medicaid population but 39 percent of its spending. The study highlights the potential benefits of the ACO model for dual eligible patients, a group for whom improved quality and coordination of care is especially important.

The PGPD was a Medicare demonstration that ran from 2005 to 2010, in which 10 participating physician group practices received bonus payments if they met quality targets and achieved lower growth in cost compared to Medicare spending for patients in their region who got most of their care from other providers. On average, the study found that the PGPD trimmed the cost of serving dual eligibles by $532 annually (after adjustment for inflation), almost five times more than the $114 per beneficiary savings generated in the general Medicare population. Spending on dual eligibles in the demonstration grew at only 60 percent of the spending rate for those in the control population.

Dartmouth’s analysis of Medicare spending for PGPD patients found that the participating health systems achieved their savings largely by reducing hospital stays. An accompanying analysis of quality indicators also showed that quality of care did not decline.

“Patients that receive both Medicare and Medicaid coverage have historically proven to be a difficult group to manage because of high illness burden, low socioeconomic status, and lack of social supports. Our results suggest that while some care management or coordination programs have failed to demonstrate savings, ACOs have the potential to improve health care and reduce spending for this population,” said Carrie H. Colla, PhD, lead author and researcher at The Dartmouth Institute for Health Policy & Clinical Practice.

The study also found significant variations in both the levels of spending and the changes in spending that were achieved by the 10 physician groups. In the 2001-2004 period before the demonstration began, average annual spending for dual eligibles ranged from $8,739 at Marshfield Clinic in Marshfield, Wis. to $17,511 at the University of Michigan Faculty Group Practice in Ann Arbor, Mich. In the subsequent five years during the demonstration, both groups achieved substantial savings for care provided to dual eligibles. Marshfield decreased annual spending by $987 per dual eligible compared to the control group and University of Michigan decreased annual spending by $2,499 per dual eligible. The Park Nicollet Clinic in Minneapolis, Minn. also experienced significant savings, decreasing annual spending by $1,610 per dual eligible compared to the control group.

Although some policy analysts have suggested that health systems with higher levels of spending may have greater opportunities to achieve savings, the evidence from this study found that both Marshfield and Park Nicollet achieved significant savings for their dual eligibles, despite relatively low levels of spending before the demonstration began.

“The current fee-for-service payment system has contributed to the fragmented, poorly coordinated care that many patients, especially those who are sick, experience every day. New payment models like ACOs are intended to encourage providers to coordinate care by offering them a share of any savings achieved when they improve care. These results indicate that when organizations really try to adapt to these new models, they can benefit their patients’ lives and their bottom lines,” said Elliott S. Fisher, MD, MPH, report author and co-principal investigator of the Dartmouth Atlas Project.

Decreased spending was achieved in large part through reductions in acute care hospitalizations, readmissions, procedures, and home health care. Reductions in spending were similar across diagnosis groups, indicating that changes in spending were due to better care management overall, rather than disease-specific interventions.

The JAMA study, “Spending Differences Associated With the Medicare Physician Group Practice Demonstration,” can be found at Additional report authors include David E. Wennberg, MD, MPH; Ellen Meara, PhD; Jonathan S. Skinner, PhD; Daniel Gottlieb, MS; Valerie A. Lewis, PhD; and Christopher M. Snyder, PhD. Funding for the study was provided by the Dartmouth Atlas Project and by support from the National Institute on Aging and the Commonwealth Fund.
The Dartmouth Atlas Project is located at The Dartmouth Institute for Health Policy & Clinical Practice and principally funded by the Robert Wood Johnson Foundation. To learn more, visit

# # #
The study used Medicare administrative data to analyze changes in spending for beneficiaries assigned to each of the 10 PGPD participants and their local control groups. Researchers used Parts A and B Medicare fee-for-service claims data for all physician groups from 2001 through 2009. For 2001-2005, researchers used 20 percent of the Medicare population, and from 2006-2009, researchers used 100 percent of Medicare beneficiaries.

About the Dartmouth Atlas Project
For more than 20 years, the Dartmouth Atlas Project has documented glaring variations in how medical resources are distributed and used in the United States. The project uses Medicare data to provide information and analysis about national, regional, and local markets, as well as hospitals and their affiliated physicians. This research has helped policymakers, the media, health care analysts and others improve their understanding of our health care system and forms the foundation for many of the ongoing efforts to improve health and health systems across America.



Annmarie Christensen
802-249-8795 (cell)

knowledge informs change