Oregon’s Medicaid Transformation – Observations on Organizational Structure and Strategy

Oregon’s Medicaid transformation has generated substantial interest among policy-makers and researchers, serving as the basis for considerable speculation (Howard, Bernell, Yoon, & Luck, 2014; McConnell et al., 2014; Pollack, 2014; Stecker, 2013, 2014). In this issue, Howard et al debate the potential for success and failure of Oregon’s Medicaid transformation to Coordinated Care Organizations (CCOs). We respond with observations based on our research that has tracked CCO development and implementation: 81 interviews with stakeholders within each CCO, substantial review of CCO documents, and monthly contact with the state. While it is too soon to declare success or failure, on the whole, CCOs have shown substantial growth in their first 18 months.

Howard et al note specific implementation challenges to CCOs. We agree that these are important considerations for CCO success. Our experience is that – 18 months into the experiment - CCOs have not entirely solved any of these problems, but they have made significant progress. In what follows, we address the challenges they point out and add additional challenges that CCOs are currently addressing.

Implementation of Electronic Health Records and Health Information Exchange

Howard et al. point to the difficulty of supporting electronic health records (EHRs) for small practices and in rural areas with low population density. While Health Information Exchange continues to be a struggle, we point to several successes thus far: In 2011, 28 percent of eligible providers had adopted certified EHRs. By September of 2013, 58 percent of eligible providers had adopted EHRs. This data does not address the interoperability of these EHRs. However, a significant number of large health systems and federally qualified health centers are on the same EHR platform, and one function known as “CareEverywhere” allows viewing hospital and outpatient notes for EPIC users (“Epic: Organizations on the Care Everywhere Network”). Furthermore, 50 local Oregon hospitals are participating in the Emergency Department Information Exchange (EDIE); care providers to develop and implement guidelines for high-utilization and special needs patients, and communicate those care coordination guidelines to other EDIE-participating facilities (“Emergency Department Information Exchange, Collective Medical Technologies”). These local steps are important because they demonstrate willingness to exchange information, and clear action steps toward sharing information, even if it is on a limited or focused scale. Smaller, independent organizations (e.g., hospitals, primary care practices) that are not, for instance on EPIC, will continue to have limited capacity to exchange information. However, the state understands the importance of a statewide Health Information Exchange and has created a Business Plan Framework through 2017 to support this endeavor and build a broader, more inclusive infrastructure for information sharing (“State of Oregon: Health Information Technology Oversight Council”).

Quality Improvement: Implementation and Measurement of CCO Performance

Quality Improvement (QI) is a continuous and formal approach to the analysis of performance and systematic efforts to improve it (“Quality Improvement”). The Agency for Healthcare Research and Quality (AHRQ) identifies the ability to understand QI approaches and how to use data and feedback to improve as a critical element of primary care practices’ ability to transform. Our research suggests that CCOs foster QI among their constituent organizations, and use data to facilitate benchmarking and to create goals and incentives for reaching QI goals.

Each CCO has a key focus on implementation of innovations to improve performance. Many are proactively collecting data from their partner organizations and feeding it back for rapid cycle improvement. The people we interview suggest that these organizations are learning from the community health assessment they are conducting and the data being collected. They use this information to develop new programs and QI efforts. This, we think, is an important developmental for these organizations and for CCOs emergence as learning health systems (Institute of Medicine, 2013).

Collaboration in governance

CCOs are designed to be local community organizations led by a Board of Directors composed of representatives from local physical, mental, behavioral, and dental health organizations. The Board of PacificSource of Central Oregon includes representation from the counties, health system, independent practice association, dental organization, and community members. This governing body was operationalized prior to the commencement of the CCOs and has become a foundation for this CCO. This CCO has had more time to develop relationships and under less pressure. While there is tension in this CCO, they have the structure and relationships to manage most hurdles. In another example of early collaboration that led to the current CCO governance structure, stakeholders from Health Share of Oregon were able to obtain Center for Medicare & Medicaid Innovation funding to implement and expand five complementary care model interventions. This early partnership brought together normally competing organizations, and board members state that there has been significant progress and cooperation. Structural problems were more pronounced in CCOs with less comprehensive and mature governance structures. Wallowa Memorial Hospital, a critical access hospital in Eastern Oregon, has no ownership stake or board representation in the Eastern Oregon Coordinated Care Organization (a newly formed organization with no history of previous partnerships and serving 12 large primarily rural and frontier counties). In southern Oregon, the Cascade Health Alliance was unable to come to terms with the county for behavioral health care. The CCO was certified following the closure of the county’s behavioral health office and the identification of a new local mental health authority. CCOs have also initiated new partnerships and opportunities for collaboration and cooperation. While challenges will continue to manifest, and the magnitude may be large and small, handled publicly and privately, all CCOs seems to be making progress identifying their key partnerships and developing relationships with those organizations.

How are CCOs and their partner organizations managing finances?

One of the largest challenges facing the CCOs is how to change the traditional payment model. Many of the former MCOs and now CCOs use capitated contracts for primary care. Few of these CCOs, however, have prior experience in alternative payment methodologies for specialists and hospital care, and this remains a barrier to payment reform. We are just beginning to see pilots of alternative payment mechanisms, including capitation with flexible spending at the primary care level, and, in a few isolated instances, fixed budgets for hospitals. CCO stakeholders recognize the fine balance between needing to change the current payment structure and upsetting current business models, and have engaged key stakeholders to develop new alternative payment methodologies for their CCO.

Primary Care Homes and CCO efforts

The Patient-Centered Primary Care Home (PCPCH) model (Oregon’s patient-centered medical home) is a CCO cornerstone. PCPCH guidelines developed through rigorous stakeholder negotiation require 10 “must pass” services including: access, accountability, quality metrics, language translation services, and health records with standardized elements which are rated on a points system. In the last state report, more than three quarters of all patients are now attributed to PCPCHs (OHA, 2014). The CCOs are putting significant effort into this measure and many are providing support to clinics in order to achieve certification. Statewide, commercial health plans have agreed to support enhanced rates for clinics with PCPCH certification. Current data indicate that spending has increased in primary care, thus potentially these additional wraparound services are being distributed to patients and thus, in our mind, PCPCH implementation has shown great success.

Providing service for behavioral and mental health

Howard et al. also specifically note the challenge of integrating behavioral health. There is still a split between traditional mental health and embedded integrated behavioral health that has not resolved, but is starting to be addressed. We have observed promising changes in this arena – some clinics have co-located mental health and primary care. However, addiction care still lags; recent reports shows that the statewide average for screening, brief intervention and treatment remains below 1% of eligible primary care patients (“Oregon Quality and Accountability”).

How will CCOs engage with community stakeholders?

The challenge to involve community stakeholders has not been seamless; Medicaid beneficiaries and health plan administrators have not been in the habit of sitting together on governing boards. Nonetheless, the focus on community has paid dividends in unexpected ways. One CCO noted that the emphasis on community accountability forced it to look at its population differently, finding, for example, that some ethnic groups seemed to stop going to the doctor when they turned 10, and didn’t appear again until child bearing age. This observation changed the ways the CCO approached care for these populations.

Can Systems Transformation be Accomplished within 3 Years and can CCOs save money?

The Oregon Health Authority (OHA) and the Centers for Medicare and Medicaid Services (CMS) have an aggressive timeline for cost savings (Howard et al). The expectations are challenging but a multi-provider demonstration project in Massachusetts achieved cost savings within a 2 year timeline (Song et al., 2011, 2012). Oregon’s Coordinated Care Organizations, moreover, have prior experience achieving cost savings with complex patients and high utilizers of the emergency department. They can expand these projects to achieve further cost savings. The state and the CCOs acknowledge the time constraints and recognize the difference between meeting goals and metrics and changing the delivery system. Current ideas for cost savings are based on relatively easy-to-measure process metrics (e.g., screening, decrease in emergency department utilization, or follow up after mental health hospitalization) rather than patient-centered outcomes that may be better measures of delivery system change. While we recognize the importance of early wins (savings), it is the broader, more enduring changes to health system delivery that are going to make a difference in the long run, after initial savings from ‘low-hanging fruit’ are accomplished.

The OHA has attempted to be very transparent in their CCO work and publishes quarterly reports with quality and access data, financial data, and progress toward reaching benchmarks. The February 2014 report encompasses data for the through November 2013 (“Oregon Quality and Accountability”.). OHA provided data for 14 of the 17 CCO incentive metrics and highlight significant improvements as well as challenges. Spending for primary care is up by more than 18 percent. Costs for surgical services, maternity, emergency department and mental health services have all decreased. Furthermore, hospital admissions for chronic illnesses such as congestive heart failure and chronic obstructive pulmonary disease have decreased by 30% and emergency department utilization has decreased by 13% (“Oregon Quality and Accountability”). Significant progress has been made on the utilization piece and cost savings may be achievable, similar to the Massachusetts demonstration project, by year 2.

How will the ACA expansion population affect the CCOs?

Following Medicaid expansion, Oregon has had an explosion of Medicaid enrollees (“Oregon Health Plan, Medicaid, and CHIP Population”). Some CCOs were unprepared and have not been able to absorb the new members. The OHA has currently reported data through November 2013, and it will be several months until we have information on changes in such measures as patient access to providers and whether the CCOs can handle this surge.

Conclusion

While it is premature to predict success or failure for the 5 year CCO experiment, our analyses nonetheless suggest that CCOs appear to be viable innovations in the delivery of healthcare. Although Howard and colleagues have presented implementation challenges, the CCOs have addressed many of these, and as current data suggest, have made significant strides in many of the key areas. We believe that the current narrative and progression of the CCOs can serve as a learning experience for other states and organizations embarking on a similar journey.