Accountable care organizations (ACOs) have become increasingly prevalent in state Medicaid programs as a mechanism to improve health care quality and control costs. To date, 10 states have launched Medicaid ACOs, 11 more states are developing Medicaid ACO models, and some states are beginning to report compelling results.
In January, a national CHCS webinar, made possible by The Commonwealth Fund, explored the early successes of three state Medicaid ACO programs with which CHCS worked early in their development: Minnesota’s Integrated Health Partnerships (IHPs); Oregon’s Coordinated Care Organizations (CCOs); and Vermont’s Medicaid Shared Savings Programs (MSSP). While each of these state ACO programs has different characteristics, all have effectively used the accountability built into their ACO model to produce significant savings and improve health care quality for more than 1.5 million Medicaid beneficiaries collectively. Notable initial results are outlined below:
|Program; Launch||# of ACOs; Population||Scope of Services||Payment Model||Select Results|
|Minnesota Integrated Health Partnerships (IHPs); 2013||21 ACOs; 460,000 covered lives||Physical health, behavioral health, and pharmacy (other services optional)||Shared savings using two tracks:|
(1) upside only; and (2) upside/ downside
|Oregon Coordinated Care Organizations; 2012||16 ACOs; approximately one million covered lives||Physical health, behavioral health, and oral health||Global budgets||
|Vermont’s Medicaid Shared Savings Programs; 2014||2 ACOs; 79,000 covered lives||Physical health (other services optional)||Shared savings using two tracks: (1) upside only; and (2) upside/ downside||
Each of these states is building on initial successes to further refine their programs. A few key priorities mentioned during the recent webinar are outlined below.
Identifying the Driver of Cost Savings
Minnesota, Oregon, and Vermont are each identifying the specific drivers of their ACO cost savings. The Oregon Health Authority credits reduced ED use as a primary driver for CCO savings. Another reported driver of cost savings for Oregon was increased enrollment in the Patient-centered Primary Care Home Program (PCPCH) through CCOs; Oregon reported $13 in savings in other services — such as specialty care, ED, and inpatient care — for every $1 increase in primary care expenditures related to the PCPCH program. Vermont is analyzing changes in health care utilization that may have contributed to ACO savings — for example a shift from inpatient to outpatient treatment. Finally, Minnesota noted that the state’s 21 IHPs have saved money in different ways and there are no common trends. Some IHPs focused on effective care coordination for critical parts of their covered population, and others addressed previously untreated behavioral health problems. Identifying cost drivers can help ACOs in targeting resources to amplify cost savings as well as identifying untapped areas for further controlling costs.
Continuing a Focus on Social Determinants of Health
States have found that ACOs can effectively address social determinants of health. Vermont’s MSSP program works collaboratively at the community level to identify local health and social needs and implement shared care plans that engage community social service providers in assessing patients’ non-medical needs. Oregon’s flexible services provisions allow for CCO payment of certain non-medical, health-related services. In addition, the Oregon legislature has funded transformation projects to allow CCOs to address enrollees’ social needs. CCOs also convene community advisory councils to engage community members in addressing social issues that may influence health. Minnesota is considering options to address its patients’ non-medical needs, and has also implemented Accountable Communities for Health: discrete, locally planned projects that are required to partner with ACOs to target particular populations and their health and social needs.
Clearly Evaluating Results
While the results outlined above are encouraging, there are caveats worth mentioning. For example, patients and providers are not randomly chosen to participate in Minnesota’s and Vermont’s ACO programs (and in Oregon’s case, almost all Medicaid members are enrolled in CCO programs), so there is no control group to more rigorously test the impact of ACOs on cost savings and quality improvements. The lack of a control group also makes it difficult to separate the impact of Medicaid ACO programs from other parallel payment and delivery system reform efforts. There are also concerns about the sustainability of savings generated under ACO models, with questions about how to keep savings “in the system” and not fully absorbed into state budgets or simply resulting in lower payment rates to ACOs and providers; this is especially relevant to upstream prevention measures that reduce downstream medical costs. Despite these issues, ACO programs continue to gain traction, and it is hard to argue with the underlying premise of redesigning health care delivery in a way that emphasizes quality, social determinants of health, value, and care coordination.
The Medicaid ACO programs in Minnesota, Oregon, and Vermont, along with many others across the country, offer promising models for addressing the main shortcomings of the American health care system: high costs and mediocre outcomes. In 2017, these fundamental challenges will continue to drive reform in the U.S. health care system, and Medicaid ACOs are continuing to evolve to meet them. As states identify what elements of their ACO programs are driving improvement, they will refine their models to better meet beneficiaries’ needs. As long as ACOs continue to produce results comparable to those seen in Minnesota, Oregon, and Vermont, we will likely see more and more states building their own versions of Medicaid ACOs.