Integrating Behavioral Health Under an ACO Global Budget: Barriers and Progress in Oregon

Financial barriers to behavioral health integration in Oregon Medicaid accountable care organizations (ACOs) limit opportunities to expand integrated care, but state and organizational opportunities exist.
Published Online: September 27, 2017
Jason Kroening-Roché, MD, MPH; Jennifer D. Hall, MPH; David C. Cameron, BA; Ruth Rowland, MA; and Deborah J. Cohen, PhD

This study describes challenges that coordinated care organizations (CCOs), a version of accountable care organizations, experienced when attempting to finance integrated care for Medicaid recipients in Oregon and the strategies they developed to address these barriers.

Study Design: Cross-case comparative study.

Methods: We conducted a cross-case comparative study of 5 diverse CCOs in Oregon. We interviewed key stakeholders: CCO leaders, practice leaders, and primary care and behavioral health clinicians. A multidisciplinary team analyzed data using an immersion-crystallization approach. Financial barriers to integrating care and strategies to address them emerged from this analysis. Findings were member-checked with a CCO integration workgroup to ensure wider applicability. 

Results: State legislation that initiated CCOs promoted integration expansion. CCOs, however, struggled to create sustainable funding mechanisms to support integration. This was due to regulatory and financial silos that persisted despite CCO global budget formation; concerns about actuarial soundness that limited reasonable, yet creative, uses of federal funds to support integration; and billing difficulties connected to licensing and documentation requirements for behavioral and mental health providers. Despite these barriers, CCOs, with the help of the state, supported expanding integrated care in primary care by using state funds to pilot test integration models and to promote alternative payment methodologies.

Conclusions: Oregon’s CCO mandate included a focus on better integrating medical and behavioral healthcare for Medicaid recipients. Despite this intention, challenges exist in the financing of integration, many of which state and federal leaders can address through payment and regulatory reform.

Am J Manag Care. 2017;23(9):e303-e309
Takeaway Points

Medicaid accountable care organizations (ACOs) face financial and regulatory barriers when integrating behavioral healthcare; however, opportunities remain for supporting and expanding integration despite these constraints. This work provides organization- and practice-level details of financial ACO barriers that have not been shown before. 
  • States can promote behavioral health integration among Medicaid ACOs by providing technical assistance to help ACOs leverage global budgets. 
  • Medicaid ACOs also must promote integration by investing in early pilot projects and internally reorganizing payment models. 
  • Consistent integration financing regulation across payers is needed.
Mental illness, substance use conditions, health behavior change, life stressors and crises, and stress-related physical symptoms—together referred to as behavioral health—are prevalent in primary care settings.1,2 Left untreated, behavioral health problems increase morbidity and healthcare utilization.3,4 Evidence suggests that patient experience and outcomes improve and costs are contained when behavioral and medical problems are addressed together.5-9

Under the Affordable Care Act, new organizations, such as accountable care organizations (ACOs), have emerged. ACOs are groups of doctors, hospitals, and other healthcare organizations that partner to develop contractual arrangements and innovative payment methods (eg, global budgets) that encourage integration and aim to achieve better quality, improved patient experience, and lower costs.10,11 Despite responsibility for behavioral health, however, fewer than 15% of ACOs have integrated behavioral health and medical care, with one-third having no formal relationship with behavioral health provider groups.12 This suggests there may be missed opportunities for ACOs to create systems to support integrated, whole-person care.13

Few studies have explored how ACOs support integration.14 Here, we examine the integration experiences of Medicaid ACOs in Oregon. We focus on Oregon because integrating behavioral health and medical care was an explicit mandate in the state legislation that created Oregon’s Medicaid ACOs, called coordinated care organizations (CCOs). Among CCOs, this mandate increased efforts and investment in integration over the past 2 or more years. We studied the challenges CCOs faced in financing integration and the strategies they developed to support integrated care.



In 2012, Oregon passed legislation creating 16 CCOs across the state through its Medicaid Section 1115 demonstration waiver. This waiver required CCOs to operate within an annual global budget, providing medical and behavioral healthcare to Medicaid beneficiaries (the state added oral health in 2014) and mandated a reduction in the Medicaid spending growth rate without a reduction in quality. Prespecified quality benchmarks were set, with upside and downside financial risk depending on whether or not CCOs met them.15,16 These features were intended to create a supportive environment for integrating behavioral and primary care, and these methods resembled regulatory mechanisms and financial incentives employed in other states.17,18


We purposively selected 5 of the 16 CCOs to participate in this study. The Oregon Health Authority (OHA) knew which CCOs were working on integration. With their help, we selected CCOs that were taking steps to integrate care for their region. We used an iterative sampling process, selecting 1 CCO, collecting and analyzing data, then using this information to inform subsequent CCO selection to ensure maximum variation in our sample and to monitor when saturation—the point at which no new findings emerged from the data—was reached.

Data Collection

Between April and October 2014, we conducted 1-on-1 semi-structured interviews with 4 to 10 key stakeholders at each CCO (n = 33). We selected stakeholders to represent a range of viewpoints on CCO activities and included CCO leaders (n = 19) and primary care and behavioral health clinicians (n = 14). A multidisciplinary research team developed the interview guide (see eAppendix [eAppendices available at]) and conducted 1-hour interviews, most of which were conducted in person. Eight interviews were conducted over the telephone to accommodate scheduling. Upon completing interviews at each CCO, we debriefed to formulate preliminary thinking and select the next CCO for study. We made additions to the interview guide when new topic areas emerged during the interviews.

Data Management

Interviews were audio-recorded, professionally transcribed, de-identified, and entered into Atlas.ti version 7.0 (Atlas.ti Scientific Software Development GmbH; Berlin, Germany) for analysis. The Institutional Review Board at Oregon Health & Science University approved this study protocol.

Data Analysis

We used an immersion-crystallization approach whereby our research team analyzed the data, without a priori presuppositions of what was important and why, and continued this process until themes or a patterned set of findings emerged.19 First, the team analyzed data for each CCO. We did this together, discussing key interview passages and, giving them a name, ultimately developed a code list. When codes and definitions were clear, and we applied them similarly during analysis (after reviewing approximately one-third of the data together), we transitioned to having 3 study team members independently code the remaining data, holding weekly meetings to ensure reliability, discuss analytical questions, and identify emerging findings.

Second, the team conducted a cross-CCO analysis, examining how themes manifested across CCOs. We developed a matrix to compare CCOs on dimensions that were emerging as important (eg, integration approach, access to psychiatry and other specialty services, challenges in financing integration). This table included characteristics of the CCO (eg, size of CCO, organizational structure, location). Finally, we prepared preliminary findings and shared them with a leadership group of clinicians and managers with hands-on experience integrating care who represented CCOs across Oregon. This allowed us to member-check and assured the transportability of findings. Stakeholders agreed that the presented findings aligned with their experience and helped us identify areas where more details were needed. We re-analyzed our data and followed up with key informants, as needed, to refine the results.


As shown in Table 1, CCOs varied in size, organizational structure, geographic location, and experience with integration. We conducted interviews with key stakeholders at each CCO. Table 2 shows the number of interviews and selected stakeholders within each CCO, with variations shaped by CCO organizational structure, size, and presence of integration infrastructure. We found that some CCOs were planning their integration efforts, while others were experimenting with fully integrated clinics and alternative payment models (APMs).

Practices within CCOs were reorganizing care delivery in several ways. As Table 3 shows, CCOs were contracting with community mental health centers (CMHCs) or other mental health organizations to embed mental health providers in regional schools; primary care clinics were bringing behavioral health clinicians (BHCs) onsite by hiring them directly or contracting with local mental health organizations; CMHCs did the same with primary care clinicians. In our interviews, stakeholders rarely described implementing specific integration models (eg, the collaborative care model); they defined behavioral health more generally, at times including mental health and substance use, and described the role of brief interventions by BHCs in primary care clinics and the delivery of basic primary care in CMHCs. With regard to access to psychiatric services, primary care stakeholders acknowledged that a small portion of their population needed these services, but psychiatric services were scarce. CCOs and practices were developing pathways to psychiatry for their patients, either through virtual connections (eg, telepsychiatry), consultation models, or by developing stronger relationships with regional mental health organizations that employed psychiatrists.

Stakeholders expressed widespread support for integration and optimism for global budgets to promote it: “What I’m hoping is that the CCO model will allow for some increased flexibility where we can just use our global budget to put the money where we think…there’ll be a payoff for the Triple Aim” (CCO 2, behavioral health director). We found, however, that CCOs’ new global budgets had not yet yielded this hoped-for flexibility to remodel care, and those CCOs and clinicians faced barriers to integration.


CCO Barriers to Integration

CCO stakeholders reported similar barriers to integrating behavioral health and medical care. These included structural financial barriers rooted in pre-CCO contracts, regulatory restrictions that affected reimbursement, and federal limitations on flexible funds, which some CCOs planned to use for integration. Table 4 summarizes these barriers, which we describe in more detail below.

PDF is available on the last page.
Adult ADHD Compendium
COPD Compendium
Dermatology Compendium
Diabetes Compendium
Hematology Compendium
Immuno-oncology Compendium
Lipids Compendium
MACRA Compendium
Neutropenia Compendium
Oncology Compendium
Pain Compendium
Reimbursement Compendium
Rheumatoid Arthritis Compendium
Know Your News
HF Compendium
Managed Care PODCAST