Comparing State ACO Efforts in Oregon and Colorado

The nation’s move from fee-for-service and the implementation of the Medicare Access and CHIP Reauthorization Act have placed more emphasis on the success of new delivery models such as accountable care organizations (ACOs). A recent paper in JAMA Internal Medicine examined the ACO programs in Colorado and Oregon to determine their impacts on spending, access, and utilization.

Oregon initiated its ACO program—moving most Medicaid enrollees into 16 Coordinated Care Organizations (CCOs)—in 2012, while Colorado created 7 Regional Care Collaborative Organizations (RCCOs) in 2011 as part of its Medicaid Accountable Care Collaborative (ACC). These 2 states represent early adoption of the Medicaid ACO model.

The Oregon program managed care within a global budget with CCOs accepting full financial risk for the patient population, while the Colorado program provided per member per month funding to the RCCOs to coordinate care and connect Medicaid enrollees with community services.

The researchers compared performance in Oregon’s CCO model with Colorado’s ACC model. They analyzed claims data from each state’s Medicaid agency for 18 months of preintervention data and 24 months of postintervention data.

“More than 2 years into their programs, both states can point to successes,” the authors determined.

Both states reported reductions in measures of standardized expenditures and utilization. But the researchers found that the Oregon model was not associated with a reduction in standardized expenditures for some services when compared with Colorado. In addition, while both states saw decreased primary care visits, they were significantly lower in Oregon than in Colorado in 2014.

Compared with Colorado, Oregon’s CCO model saw improvements in 3 of 4 Healthcare Effectiveness Data and Information Set access measures, as well as reductions in both avoidable emergency department visits and preventable acute hospital admissions. However, the CCO model was not associated with significant improvements in 3 of 4 measures of low-value care compared with Colorado.

“Compared with Colorado, Oregon experienced improvements in some access and quality measures, but did not generate savings that might be anticipated with its ambitious reform model and the $1.9 billion federal investment to support the CCO transformation,” the authors wrote.

They determined that Colorado’s ACC model might represent a more promising delivery system reform option for other states to adopt.

“The incentives in most existing alternative payment models, including ACOs, are commonly considered insufficient to result in behavior change,” Carrie H. Colla, PhD, and Elliott S. Fisher, MD, MPH, wrote in an accompanying commentary. “However, the Colorado study suggests that strong incentives may not be necessary. Rather, the Colorado medical home model improved value through supporting providers with coaching, connecting members with nonmedical services, and providing feedback on costs, utilization, and outcomes.”