The Impact Of Medicare ACOs On Improving Integration And Coordination Of Physical And Behavioral Health Care

Accountable care organizations (ACOs) are groups of doctors, hospitals, and other providers that come together voluntarily to provide high-quality care to a defined population of patients and that are collectively held financially responsible for those patients’ total cost of care. ACOs have proliferated across the United States, spurred by new commercial payer contracting options and payment models sponsored by the Centers for Medicare and Medicaid Services (CMS). The ACO model holds the promise of reducing costs while improving health outcomes and the quality of care by realigning payment incentives to focus on the total cost of care instead of per service payments.

The Medicare Pioneer ACO Model and the Medicare Shared Savings Program began in 2012. ^1,2 ACOs participating in the Medicare Shared Savings Program that had limited access to capital and met other criteria (these ACOs were typically smaller practices or rural providers) were also eligible to apply to participate in the Advance Payment ACO Model, in which they would receive advance payments to defray start-up costs—payments that would be repaid from future shared savings that they would receive.

Medicare ACOs can be structured in a variety of ways, including integrated delivery systems, groups of independent practices, multispecialty group practices, partnerships between hospitals and providers, and hospital-led organizations. The ACOs are not required to include specific specialty provider types, such as behavioral health care providers. However, to receive shared savings, they must reduce the total cost of care for their attributed population while meeting performance standards on thirty-three quality measures, such as measures of patient or caregiver experience, care coordination, patient safety, preventive health care, and managing at-risk populations (for example, patients with diabetes, hypertension, or heart disease). ³ Because of the impact that behavioral health has on total cost of care, Medicare ACOs could be expected to include behavioral health in their efforts. ⁴

One important aspect of managing the total cost of care is improving care coordination and treatment for Medicare beneficiaries with mental or substance use disorders. ⁵ Research has shown that people who are high utilizers of health care services and those with chronic diseases such as diabetes, cardiovascular disease, pulmonary disease, and arthritis have a higher prevalence of behavioral health conditions. ⁶ Total costs of care for patients with either diabetes or heart failure and comorbid depression are 4.5 times higher than for those without mental health comorbidities. ³ Behavioral health conditions in patients with chronic medical comorbidities have also been associated with poorer adherence to treatment, greater disability, and poorer quality of care, compared to similar patients without behavioral health conditions. ^7–9

Researchers have found that collaborative and integrated care models are effective in responding to the behavioral and physical health care needs of people with comorbid conditions. ^10–12 Examples include models that integrate mental health providers into the primary care setting ¹³ and Screening, Brief Intervention, and Referral to Treatment (SBIRT) for substance use disorders. ¹⁴ These models have not been adopted as rapidly as one might expect, given their effectiveness. ¹⁵ Potential barriers to implementation in fee-for-service (FFS) settings include the models’ complexity, adaptability, and cost. ¹⁶

Previous research on ACOs based on a 2013 survey suggests that few ACOs have moved beyond the traditional and fragmented arrangements common under FFS reimbursement. ¹⁷ However, the results of this research were based on a limited time period, and organizations that are embracing the ACO model and other population-based payment models are rapidly adapting their initiatives and capabilities. ¹⁸

In this study we examined qualitative data from multiple interviews with a variety of executives, administrative personnel, and providers at Medicare ACOs from 2012 through 2015 to determine whether and how these organizations expanded connections between primary care and behavioral health care. Specifically, we examined the extent to which the ACOs recognized and focused on behavioral health as an important contributor to improving quality of care and generating savings, the types of approaches ACOs used to address behavioral health, and the primary challenges they faced when trying to implement improvements in behavioral health care.

Our results will help providers and policy makers understand whether an ACO model leads to improvement in behavioral health care and will help ACO officials understand how their peers have identified and addressed behavioral health needs within the framework of the ACO delivery model.

Study Data And Methods

We used data from quarterly assessments of and site visits at ninety ACOs that an evaluation team conducted under a CMS contract to independently evaluate the Pioneer and Advance Payment ACO models.

Quarterly Assessments

Between December 2012 and June 2015, an evaluation team consisting of more than twenty qualitative researchers conducted up to nine semistructured quarterly telephone interviews with personnel at each of ninety Medicare ACOs. To maintain continuity and communication, two evaluation team members were assigned to each ACO to conduct the quarterly interviews and site visits. CMS’s Center for Medicare and Medicaid Innovation, which manages Medicare’s ACO program, provided ACO contact information. Before each interview the ACOs were informed of the topic or topics to be covered, so they could invite the most appropriate personnel to answer the questions. As a result, the number of individuals on each call varied within and across ACOs.

Each quarterly interview focused on one or two topics (such as care coordination, ACO structure, and data analytics) related to efforts to improve care and reduce costs. Many of these discussions included behavioral health. Specific questions about ACO activities related to behavioral health were included in a final interview. Sixty-nine of the ninety ACOs had final interviews available for analysis. We included all available quarterly interviews in our analysis even if an ACO had withdrawn from the Medicare program or for some other reason did not have a final interview available for analysis.

Site Visits

Between September 2013 and July 2014, the evaluation team conducted two-day site visits with seventy-nine of the ACOs. The itinerary for each site visit was organized and coordinated in conjunction with the ACO and typically included interviews with executive, administrative, and provider staff members, including physicians. After each site visit, team members summarized the findings and observations in a debriefing document. Eleven ACOs either had left their ACO program before September 2013 or did not agree to participate in a site visit.

Cohort Description

The initial evaluation cohort included thirty-two Pioneer ACOs, twenty-three Shared Savings Program ACOs, and thirty-five Advance Payment ACOs ( Exhibit 1 ). Twenty-one ACOs discontinued participation in their program before the last quarterly assessment, or that assessment was not available for analysis. For the remaining sixty-nine ACOs that completed the final quarterly assessment, we categorized them as having a behavioral health initiative if they mentioned performing any of the following activities: integrating behavioral health providers into or collocating them with primary care practices; integrating social workers into care coordination teams; using social workers to manage behavioral health needs; and strengthening or expanding their behavioral health referral networks, processes, or both. We then summarized ACO characteristics by whether or not the ACO had a behavioral health initiative.

Qualitative Analysis

The evaluation team uploaded content from the records of the quarterly assessments and site visits into a qualitative analytic software program (Dedoose). Five evaluation team members reviewed this content and coded as behavioral health any excerpts related to behavioral health initiatives or strategies. The excerpts were downloaded into an Excel file for further categorization. Two evaluation team members developed initial repeating themes based on a review of the excerpts coded as behavioral health from the first three quarterly assessments of the Pioneer ACOs. The same two individuals further developed the behavioral health themes based on a review of the behavioral health excerpts after all material had been coded.

Limitations

Our study had several limitations. First, the same team members did not conduct all of the interviews. Although we used semistructured interview guides, interviewers varied in the breadth of their questioning about behavioral health issues, and respondents varied in the comprehensiveness of the information they shared.

Second, the accuracy of the answers depended on the knowledge of the individuals at the ACOs who were participating in the interviews.

Study Results

Almost all ACO personnel recognized the contribution of behavioral health disorders to beneficiaries’ utilization and spending, particularly once the personnel focused on understanding the factors that contributed to making some people high-cost beneficiaries in their ACO. At many ACOs, care coordination teams recognized that a greater percentage of their high-risk and high-cost beneficiaries had complex behavioral health and psychosocial needs, compared to beneficiaries who were not high risk and high cost. Furthermore, ACOs recognized the connection between behavioral health and high-cost beneficiaries as they examined the reasons for repeat hospitalizations, repeated use of the emergency department, and longer hospital stays than expected.

Of the sixty-nine ACOs that completed the final interview, forty-three had developed at least one behavioral health initiative to meet their beneficiaries’ needs ( Exhibit 2 ). More Pioneer and Shared Savings Program ACOs developed at least one approach than did Advance Payment ACOs. In addition, the ACO structure most likely to have addressed behavioral health needs was the integrated delivery system. These systems include, and coordinate the work of, multiple specialties and both outpatient and inpatient treatment settings.

Focus On Behavioral Health Issues

Most ACOs initiated or expanded programs to provide behavioral health care for their beneficiaries and to improve the coordination of that care between primary care and behavioral health care providers. Approaches ranged from implementing integrated care models to improving relationships with behavioral health care providers outside the ACO ( Exhibit 3 ). Many ACOs implemented multiple approaches to address the differing needs of their beneficiaries and providers. These efforts were often embedded in larger delivery transformation efforts, such as a reorganization of the providers participating in an ACO to become patient-centered medical homes with an increased emphasis on care coordination. Often there was variation within an organization regarding behavioral health access, with organizations reporting integrated care at some sites and a lack of providers at others.

Many behavioral health initiatives evolved over time as the availability of staff or other resources expanded or contracted. Some organizations implemented these initiatives as part of their larger ACO initiatives or in conjunction with other ongoing initiatives. Other ACOs recognized a need to address behavioral health as they worked with their high-cost beneficiaries and expanded their referral networks. Finally, some ACOs were unable to maintain their initial investment in social worker time to assist with behavioral health coordination and short-term support because of lack of funding or staff availability. In the end, many ACOs attempted to address behavioral health via a trial-and-error approach that aligned the resources of the organization and community with the perceived behavioral health needs of their beneficiaries.

Collocated Care

Many ACO personnel discussed having integrated behavioral health and primary care. Several had implemented a model based loosely on chronic care models or the Improving Mood—Promoting Access to Collaborative Treatment (IMPACT) model. ¹⁹ Most ACOs used licensed clinical social workers instead of nurse practitioners to treat mental, behavioral, and emotional issues, especially depression. Several ACOs focused on developing programs collocated with primary care providers to treat depression in individuals with chronic physical conditions such as diabetes or cardiovascular disease. Many of these programs were developed and funded concurrently with the ACO itself, through outside grants or initiatives.

Other ACOs placed behavioral health care providers (usually licensed clinical social workers) in primary care sites to provide assistance with initial referrals for behavioral health services or other short-term needs. Some ACOs placed these providers in clinics specifically developed for patients with complex needs. Primary care physicians and care coordinators referred patients to these clinics, where both their physical and behavioral health needs could be treated. At least one ACO had such an integrated clinic for patients with complex needs and an integrated team that could visit patients at home.

Several ACOs enhanced the behavioral health programs developed before their formation as ACOs to integrate them more fully into the primary care teams. One ACO leader described the process of incremental change: “We have embedded mental health providers in the primary care practices. They had been there preceding Medicare shared savings, but it was more of a geographic cohabitation. Now it’s becoming more integrated into the actual team and within the context of the medical home.”

Some ACOs also used federally qualified health centers to provide integrated care. Many ACO respondents discussed how the previous inclusion of behavioral health care in some capacity in these organizations placed the ACOs that included federally qualified health centers ahead of larger ACOs that were struggling to develop models to address behavioral health care. A handful of ACOs included as a partner a significant outpatient mental health facility, such as a community mental health center or other large stand-alone mental health provider. At least two ACOs took the opposite tack and embedded primary care services into their mental health clinics.

Increased Access To Social Workers

In contrast to using licensed clinical social workers as providers collocated with primary care teams, many ACOs hired social workers to support their medical care coordination teams or to serve as independent centralized resources for both short-term behavioral health care and long-term coordination of referrals for mental health treatment. Medical care coordination teams often included pharmacists, licensed clinical social workers (or, less frequently, other social workers), and community resource specialists in addition to nurse care coordinators. The coordinators referred patients with significant mental health issues to social workers for assistance in resolving psychosocial issues, providing short-term behavioral health services, or coordinating long-term mental health treatment.

In other ACOs, social workers played a key role in assisting with behavioral health care needs without being explicitly included in the care management team. One ACO had initially placed a social worker in one of its primary care clinics, but demand was so high that it centralized the social worker so that he or she was available to other primary care clinics. Other ACOs had centralized social work teams. One ACO used social workers to follow up with patients identified through depression screening.

Referral Networks

Multiple ACOs adjusted their referral networks to better serve beneficiaries with behavioral health needs by improving connections to community resources, partnering with a behavioral health facility to improve access to care, and reorganizing internal behavioral health resources to improve access to and coordination with primary care providers. Several ACOs expanded their behavioral health networks as part of their care coordination efforts. Care coordination team members catalogued community resources and reached out to them, often providing a bridge to connect beneficiaries to these resources.

Other ACOs created partnerships with behavioral health organizations that varied from formal contractual relationships to informal understandings. As an official at one Pioneer ACO put it: “We don’t have a robust mental health program, so we have a good referral relationship to a behavioral health center. The behavioral health center outpatient people come to ACO management meetings but are not contractually affiliated [with the ACO].”

Another ACO had a social worker at its main primary care clinic but had backup psychiatric support through a partnership with a mental health organization. Many ACOs also referred beneficiaries to home health agencies and visiting nurse associations for behavioral health and social work services. Several ACOs explored the provision of mental health care via videoconferences for outpatients to expand access, although few had developed the ability to provide such services.

Several ACOs also shifted their internal resources to improve referral-based behavioral health services. One Shared Savings Program ACO developed a mental health “center of excellence” to which primary care physicians could refer patients when their care needs exceeded what could be provided in the clinic. This “center of excellence” housed a behavioral health team that could address the complex care needs of these patients and coordinate with primary care physicians in the same organization.

One Pioneer ACO leader reported developing a type of concierge model: “The psychiatrist consults [with the primary care provider] and provides a recommendation on medication management, and then the patient goes back to their primary care physician for ongoing management and has orders for interim consults with the psychiatrist as needed.”

Approaches To Behavioral Health Screening

Because depression screening is one of the quality metrics that CMS collects, many ACOs enhanced the depression or behavioral health screenings in their primary care settings. ACOs varied in how they handled patients who screened positive for mental illness or substance use disorders during these visits.

Some ACOs developed pathways for people with positive depression screens that included prompts showing the provider how to respond. Others referred such people to social workers within the health system or collocated with primary care providers for treatment or referral assistance. Some ACOs taught their primary care providers how to screen for depression, educated them about the value of managing patients with behavioral health issues, or both. Others were resistant to screening and noted that physicians could identify mental illness without using a screening tool. Personnel at one ACO explained that routine depression screens were not a substitute for an existing relationship between the beneficiary and his or her primary care provider.

Organizations Not Focusing On Behavioral Health Issues

Some ACOs acknowledged the existence of unmet health care needs for Medicare beneficiaries with behavioral health issues but had not yet addressed this gap in care. We classified these ACOs into three groups.

In the first group, behavioral health was a priority that had not yet been addressed. Some ACOs planned to implement specific programs, such as telepsychiatry initiatives and pilot programs to integrate licensed clinical social workers into their care model. Others were in earlier stages of formulating their strategies or noted that behavioral health had been replaced by other priorities. While believing that focusing on behavioral health would benefit their organizations, personnel at two ACOs said that they did not have resources to try new things because they had lost money on their ACO initiative or did not have a financing model in place to make necessary changes.

ACOs in the second group felt that behavioral health care was better addressed at the practice level than at the ACO level. These ACOs supported the traditional approach, in which each practice managed beneficiaries with behavioral health issues. As one ACO official explained, each practice has specific knowledge about the local resources available to beneficiaries in its area.

In the third group was a handful of ACOs that did not consider behavioral health to be a priority. Two of these ACOs felt that mental health was better addressed at the community level than at the ACO level. One ACO official noted that physicians did not have time to explore behavioral health issues with patients. Another felt that there was more leverage to improve behavioral health care under the Medicare Advantage model than under the ACO model. And still another noted that the focus of his ACO had been on “other things” and that “most primary care physicians are referring [patients] out for different mental illnesses.”

Some ACO personnel noted that their ACO’s care management programs would help improve the physical health care of their beneficiaries with behavioral health needs even without the resources of a social worker or other behavioral health program. Finally, some noted that they were not doing anything beyond trying to improve depression screening rates—which, as noted above, is a quality measure that ACOs are required to report to Medicare.

Challenges

Both ACOs that have improved their integration of behavioral health care services into primary care services or the coordination of both services and ACOs that have not yet addressed these issues described facing similar challenges, some of which are listed in Exhibit 4 .

Lack of behavioral health professionals was one frequently cited challenge that referred both to a scarcity of providers (either within the ACO or within the surrounding community, to whom patients could be referred) and to geographic constraints (such as being located in a rural or remote area). Many ACOs did not have enough sources of community referrals or a labor pool from which to hire behavioral health specialist staff members, particularly psychiatrists and professionals with expertise in substance use disorders.

Some ACO personnel speculated that the lack of behavioral health professionals was attributable to poor reimbursement rates, with some noting low Medicaid reimbursement. Others observed that some types of licensed behavioral health providers could not bill to Medicare, while still others reported that many behavioral health care providers did not see Medicare patients because of the program’s historically limited behavioral health coverage.

Developing a sustainable funding model for behavioral health services in an FFS-based reimbursement system, even in the context of an ACO, was challenging for many organizations. Many ACOs said that the FFS system and insurance coverage that carved out behavioral health helped drive the separation between behavioral and physical health care. One Pioneer ACO official stated: “The value of behavioral health is high under a capitated system but very low under FFS. That’s a positive of the ACO model, that it will reintegrate behavioral health.”

Generally, ACOs funded behavioral health and care coordination in similar ways—that is, by depending on outside funding or funding from their profits, instead of anticipating that these providers would cover their salaries through billing.

ACO personnel also described challenges related to sharing mental and substance use disorder data, because these data require extra security protection. Many interviewees said they did not receive reliable behavioral health data for their beneficiaries from CMS because of the suppression of data that include a substance use disorder diagnosis or related procedure. As a result, ACOs could not perform the same data analytics to understand the needs of this population that they could perform for other chronic diseases, nor could they use claims data to identify beneficiaries for additional outreach. ACO personnel also described how privacy restrictions limited the use of internal electronic health record data to identify people with behavioral health needs and hindered the coordination of care between separate behavioral and physical health care sites.

Some ACO personnel described resistance to discussing mental health on the part of beneficiaries and providers. These challenges included the cultural stigma that reduced providers’ ability to successfully refer beneficiaries to behavioral health care providers or otherwise engage them in treatment; providers’ resistance to screening for depression or mental health without clear pathways for behavioral health treatment or referral mechanisms; and resistance from psychiatrists who feared that the complexity of their patients’ needs would increase if care for depression was provided in primary care settings that used integrated social workers for less complex patients.

Organizations More Focused On Mental Illness Than On Substance Use Disorders

In interviews and site visits, it became clear that ACOs focused more on mental illness than on substance use disorders. In their responses to questions about behavioral health, some ACO representatives commented only on what they were doing for patients with depression, anxiety, and other mental illnesses. Others explicitly said that they were focusing on mental illness and not on substance use disorder treatment, despite acknowledging that the latter was an important issue.

One interviewee said that finding appropriate language to identify and discuss substance use disorders required training. Another noted a significant lack of substance use disorder providers. Two mentioned that their ACOs were doing nothing for substance use disorder treatment beyond using a patient contract when prescribing Schedule II drugs (those categorized by the Drug Enforcement Administration as having a high potential for abuse that may lead to psychological or physical dependence).

Some ACO representatives mentioned looking for prescription drug abuse or reviewing claims for overuse or misuse of drugs susceptible to addiction. One mentioned teaching motivational interviewing ²⁰ to providers to use along with substance use disorder screenings as a way to motivate change.

Discussion

Representatives of Medicare ACOs recognized that behavioral health conditions play an important role in making some beneficiaries high cost, and most of the representatives we interviewed said that their ACOs had implemented changes to better address behavioral health care needs. The two most common models were integrating behavioral health care providers (usually licensed clinical or bachelor’s-level social workers) into primary care and using social workers in medical care coordination teams to provide time-limited support and assistance with more permanent referrals when needed.

Most ACOs pursued more than one approach, and the level of integration varied across the organizations’ sites. Most ACOs provided integrated care on a limited basis, either to selected patients (such as those with specific physical health conditions and depression) or in selected locations (such as in larger sites or those focused on beneficiaries with complex needs). At the same time, most ACOs supported other providers through enhanced clinical pathways and through social workers connected to the care coordination team, who helped individuals needing ongoing or long-term care through enhanced referral networks. In this way, ACOs addressed both individuals who could be treated in primary care clinics and those who required more complex care than is available in such clinics.

Many of the behavioral health programs being developed and implemented either existed before the ACO or were part of concurrent initiatives of the organization that dovetailed with the ACO initiatives. Many ACOs had expanded existing programs or were in the midst of large-scale delivery system changes (for example, becoming patient-centered medical homes) that allowed for enhanced care coordination. These programs are rapidly evolving.

The funding for many such models (particularly the integrated care models) depended on grants and organizational discretionary funds, which are not readily available in many organizational structures—particularly Advance Payment ACOs. This highlights one of the main challenges in trying to bridge the gaps between FFS and population-based payment approaches. The ACO payment model represents a hybrid between the two systems, and continued uncertainty about receiving shared savings poses challenges for sustainable funding. In addition to the ACO payment approach, multiple initiatives are exploring ways to provide sustained funding for integrated care models, such as bundled payments. ²¹

Many ACOs have not made behavioral health a priority or feel overwhelmed and unsure about how to improve behavioral health services. The significant challenges described by ACOs that were related to access to behavioral health providers and to the integration of behavioral health care providers into primary care settings have resulted from a long history of fragmented care and will take time to resolve. The fact that only a few ACOs had made efforts to improve care for people with substance use disorders suggests that the challenges associated with treating this population may be even more complex than those involved in improving care for people with mental health disorders.

Despite these limitations, there appears to be increasing interest among ACOs in better coordinating behavioral health with and integrating it into primary care. Many representatives of well-established ACOs said that they sensed a change within their organizations toward trying to rediscover experiences from a time when behavioral health was better integrated into primary care than the current FFS system permits. It may not be easy to recover the resources and knowledge that existed then, but we got a clear sense from interviewees that as payment models from public and private payers continue to move away from FFS and toward capitated payments, integration of behavioral health care services will return to the mainstream.

Conclusion

ACOs are increasingly recognizing the importance of addressing behavioral health care needs to help control utilization and spending, but they have had mixed success in trying to integrate behavioral health into and coordinate it with primary care. Many organizations have collocated behavioral health providers with primary care providers and improved coordination between the two groups as they reorganized themselves into ACOs and adopted new payment models and delivery structures. Many others have created mechanisms to provide short-term behavioral health support while increasing the strength of their referral networks for long-term treatment.

Significant challenges remain, such as those related to provider access, sustainable financing models, data sharing, and resistance to treatment by both providers and patients. However, the ongoing shift toward capitated payments is driving a new era of coordination and integration between physical and behavioral health that will expand the models for effective and improved health care to address all health needs.

ACKNOWLEDGMENTS

Preliminary results from this study were presented at the AcademyHealth Annual Research Meeting in San Diego, California, June 8–10, 2014. The analysis was performed under subcontract to L&M Policy Research, LLC (Contract No. HHSM-500-2011-000191/T0002 from the Center for Medicare and Medicaid Innovation of the Centers for Medicare and Medicaid Services [CMS]). The authors acknowledge the research staff at the following firms for conducting the interviews and site visits: Abt Associates; Avalere Health; L&M Policy Research, LLC; Social and Scientific Systems; and Truven Health Analytics. In addition, the authors thank Lisa Green from L&M Policy Research and David Nyweide from CMS for helpful comments and reviews of the manuscript. Finally, the authors thank Linda Lee and Paige Jackson for editorial assistance.

NOTES

• 1 CMS.gov . Pioneer ACO Model [Internet]. Baltimore (MD) : Centers for Medicare and Medicaid Services ; [last updated 2016 May 18; cited 2016 May 27 ]. Available from: https://innovation.cms.gov/initiatives/Pioneer-ACO-Model/index.html Google Scholar
• 2 CMS.gov . First accountable care organizations under the Medicare Shared Savings Program [Internet]. Baltimore (MD) : Centers for Medicare and Medicaid Services ; 2012 Apr 10 [cited 2016 May 27 ]. (Fact sheet). Available from: https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2012-Fact-sheets-items/2012-04-10.html Google Scholar
• 3 Centers for Medicare and Medicaid Services . Table: 33 ACO quality measures [Internet]. Baltimore(MD) : Centers for Medicare and Medicaid Services ; [cited 2016 May 27 ]. Available from: https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/sharedsavingsprogram/Downloads/ACO-Shared-Savings-Program-Quality-Measures.pdf Google Scholar
• 4 O’Donnell AN , Williams BC , Eisenberg D , Kilbourne AM . Mental health in ACOs: missed opportunities and low-hanging fruit . Am J Manag Care . 2013 ; 19 ( 3 ): 180 – 4 . Medline, Google Scholar
• 5 Maust DT , Oslin DW , Marcus SC . Mental health care in the accountable care organization . Psychiatr Serv . 2013 ; 64 ( 9 ): 908 – 10 . Crossref, Medline, Google Scholar
• 6 Egede LE , Zheng D , Simpson K . Comorbid depression is associated with increased health care use and expenditures in individuals with diabetes . Diabetes Care . 2002 ; 25 ( 3 ): 464 – 70 . Crossref, Medline, Google Scholar
• 7 Sayers SL , Hanrahan N , Kutney A , Clarke SP , Reis BF , Reigel B . Psychiatric comorbidity and greater hospitalization risk, longer length of stay, and higher hospitalization costs in older adults with heart failure . J Am Geriatr Soc . 2007 ; 55 ( 10 ): 1585 – 91 . Crossref, Medline, Google Scholar
• 8 Druss BG , Zhao L , Cummings JR , Shim RS , Rust GS , Marcus SC . Mental comorbidity and quality of diabetes care under Medicaid: a 50-state analysis . Med Care . 2012 ; 50 ( 5 ): 428 – 33 . Crossref, Medline, Google Scholar
• 9 Reeves WC , Strine TW , Pratt LA , Thomson W , Ahluwalia I , Dhingra SS , et al. Mental illness surveillance among adults in the United States . MMWR Suppl . 2011 ; 60 ( 3 ): 1 – 29 . Medline, Google Scholar
• 10 Department of Health and Human Services . Mental health: a report of the surgeon general [Internet]. Washington (DC) : HHS ; 1999 ; [cited 2016 May 27 ]. Available from: https://profiles.nlm.nih.gov/ps/access/NNBBHS.pdf Google Scholar
• 11 Drake RE , Mercer-McFadden C , Mueser KT , McHugo GJ , Bond GR . Review of integrated mental health and substance abuse treatment for patients with dual disorders . Schizophr Bull . 1998 ; 24 ( 4 ): 589 – 608 . Crossref, Medline, Google Scholar
• 12 Woltmann E , Grogan-Kaylor A , Peron B , Georges H , Kilbourne AM , Bauer MS . Comparative effectiveness of collaborative chronic care models for mental health conditions across primary, specialty, and behavioral health care settings: systematic review and meta-analysis . Am J Psychiatry . 2012 ; 169 ( 8 ): 790 – 804 . Crossref, Medline, Google Scholar
• 13 Butler M , Kane RL , McAlpine D , Kathol RG , Fu SS , Hagedorn H , et al. Integration of mental health/substance abuse and primary care [Internet]. Rockville (MD) : Agency for Healthcare Research and Quality ; 2008 Oct [cited 2016 May 27 ]. (AHRQ Publication No. 09-E003). Available from: http://www.ahrq.gov/sites/default/files/wysiwyg/research/findings/evidence-based-reports/mhsapc-evidence-report.pdf Google Scholar
• 14 Babor TF , McRee BG , Kassebaum PA , Grimaldi PL , Ahmed K , Bray J . Screening, Brief Intervention, and Referral to Treatment (SBIRT): toward a public health approach to the management of substance abuse . Subst Abus . 2007 ; 28 ( 3 ): 7 – 30 . Crossref, Medline, Google Scholar
• 15 Goldman HH , Ganju V , Drake RE , Gorman P , Hogan M , Hyde PS , et al. Policy implications for implementing evidence-based practices . Psychiatr Serv . 2014 ; 52 ( 12 ): 1591 – 7 . Crossref, Google Scholar
• 16 Briand C , Menear M . Implementing a continuum of evidence-based psychosocial interventions for people with severe mental illness: part 2—review of critical implementation issues . Can J Psychiatry . 2014 ; 59 ( 4 ): 187 – 95 . Crossref, Medline, Google Scholar
• 17 Lewis VA , Colla CH , Tierney K , Van Citters AD , Fisher ES , Maera E . Few ACOs pursue innovative models that integrate care for mental illness and substance abuse with primary care . Health Aff (Millwood) . 2014 ; 33 ( 10 ): 1808 – 16 . Go to the article, Google Scholar
• 18 Houston R , McGinnis T . Accountable care organizations: looking back and moving forward [Internet]. Hamilton (NJ) : Center for Health Care Strategies ; 2016 Jan [cited 2016 May 27 ]. Available from: http://www.chcs.org/media/ACOs-Looking-Back-and-Moving-Forward.pdf Google Scholar
• 19 AIMS Center . IMPACT: Improving Mood—Promoting Access to Collaborative Treatment [Internet]. Seattle (WA) : University of Washington ; c 2016 [cited 2016 May 27 ]. Available from: http://aims.uw.edu/impact-improving-mood-promoting-access-collaborative-treatment/ Google Scholar
• 20 Miller WR , Rollnick S . Motivational interviewing: helping people change . 3rd ed . New York (NY) : Guilford Press ; 2013 . Google Scholar
• 21 O’Donnell AN , Williams M , Kilbourne AM . Overcoming roadblocks: current and emerging reimbursement strategies for integrated mental health services in primary care . J Gen Intern Med . 2013 ; 28 ( 12 ): 1667 – 72 . Crossref, Medline, Google Scholar