Neftali Serrano, PsyD
The integration of primary care and mental health services has developed significantly over the last 15 years. Service delivery models have been developed and tested, the literature base has grown and now thousands of mental health professionals self-identify as working in some form of integrated care setting. Here we will review the latest developments in the field and its future directions.
There are three primary models of care that have emerged in the last decade although implementation of the models varies significantly site-to-site. These are the Collaborative Care model, the Primary Care Behavioral Health model and the SBIRT model. While implementation of these models varies for the purposes of this article these models will serve as umbrella models to describe different iterations they can represent in actual practice.
The Collaborative Care model (Katon, et al., 2012) emerged in response to the sub-par management of depression in primary care (often referred to as ‘treatment as usual, or TAU’) and the recognition that most of depression care in the United States occurs in primary care. This model focuses on system redesign utilizing screening protocols, a care manager, a consulting psychiatrist, and sometimes co-located mental health support. The end goal of the model is to improve the identification of patients with depression in the population, improve the tracking of these patients' outcomes via use of specified contact/visit algorithm and to provide primary care clinicians with support from a consulting psychiatrist. This model has emerged primarily from schools of psychiatry and although protocols often do provide for limited behavioral health support for patients in the form of motivational interviewing or other manualized treatments, the main intervention is medication.
The Primary Care Behavioral Health model (Robinson, et al., 2007) emerged in response to the plethora of behavioral issues that present in primary care and the barriers patients and clinicians often face in accessing mental health services in the United States. This model places a generalist mental health professional, often called a Behavioral Health Consultant, to work alongside a primary care clinician to provide curbside consultation to the clinician and immediate or near-immediate intervention to the patient. The generalist nature of the model means that Behavioral Health Consultants work with the full range of mental health conditions that patients may present with and behaviorally-related medical issues, such as difficulties with diabetes management. Key end-goals of the model include improving access to mental health care, often called population penetration, in addition to improving medical provider satisfaction and patient outcomes. This model has emerged primarily from practice-level innovations spurred by a disparate group of entities from Federally Qualified Health Centers to the Department of Defense.
The SBIRT model (Screening, Brief Intervention & Referral to Treatment) (Barbosa, et al., 2015) emerged in response to the high rates of substance abuse issues that present in primary care and the difficulties with accessing substance abuse specialty care. SBIRT also focuses on system redesign in the form of systematic screening and includes brief intervention, either from a health coach or a licensed professional who is co-located with the primary care clinic. Patients who are deemed too complex are referred to local providers for specialty care. The key goal of this model is to engage patients with mild to moderate substance abuse issues (sometimes including tobacco abuse) and to identify and refer those with higher acuity addictions to specialty clinics. This model has proliferated as a result of substantial federal, state and local grants and research trials and as such has emerged primarily from university centers.
These three models have different end goals but are not mutually exclusive. The Collaborative care model and the SBIRT model are primarily focused on one disease condition, although recent efforts have been made to expand the scope of the conditions addressed. The Primary Care Behavioral Health model is focused on influencing treatment in the primary care clinic across a variety of behavioral health categories. Several models may run concurrently, for example, when the depression protocols and registries of the Collaborative Care model are combined with the immediate generalized support provided by a Behavioral Health Consultant or when a Behavioral Health Consultant also operates using an SBIRT protocol. Proponents of the Collaborative Care and SBIRT models tout their well-researched outcomes that demonstrate improved depression scores and screening rates relative to usual care whereas critics of the model decry its lack of impact across most of the issues that patients bring to primary care. Proponents of the Primary Care Behavioral Health model tout its high acceptability rate from patients and primary care clinicians but critics point out the lack of a robust literature base relative to patient outcomes. Sites tend to adopt models that meet their particular needs and sometimes blend aspects of each model.
Often the decision about which model to pursue depends on the talent available to a site. The new breed of integrated care professionals include care managers, nurses, consulting psychiatrists, and re-specialized mental health professionals. Care managers manage patient registries for tracking purposes and sometimes provide brief interventions either in person or over the phone. These care managers can sometimes be mental health professionals at the bachelor’s or master’s level or are sometimes nurses. Consulting psychiatrists provide consultation in written and/or verbal fashion to primary care clinicians and sometimes provide face-to-face consultation to patients with complex conditions. Behavioral Health Consultants are either master’s or doctoral level mental health clinicians who provide the curbside consultation and immediate intervention to patients. SBIRT can be administered with a licensed mental health professional or a bachelor’s level health coach.
There are a variety of other iterations or attempts at adapting versions of these models but these have emerged as the basic frameworks from which integration efforts currently emerge.
The literature supporting each of the models has grown substantially in the last decade. The Collaborative Care and SBIRT models, driven primarily via academic centers, have received the most attention in the literature. The Collaborative Care model has demonstrated improved outcomes for depression compared to usual care. Recent data support a connection between improved medical outcomes for patients with cardiac conditions and diabetes who have comorbid depression (Huffman, et al., 2011; Bogner, et al., 2012). Given the robust findings with regard to depression management, much of the literature is now focused on measuring potential impact on medical outcomes (Multicondition Collaborative Care) and assessing the economic impact of the model (Katon, et al., 2012). Studies have shown that Collaborative Care does increase costs associated with delivering the intervention but has the potential for cost offset in the form of reduced total medical cost expenditures.
The SBIRT model has demonstrated improved identification of patients in primary care and emergency department settings with substance abuse problems along with reductions in risky substance abuse. Recent literature has focused on identifying sub-populations which could benefit from SBIRT screening such as OB/GYN patients and assessing the cost of the intervention relative to cost offset to payers, particularly large payers like Medicaid (Barbosa, et al., 2015; Estee, et al., 2010). SBIRT adherents tout its relatively low cost and high potential for offset given that the screening and intervention can be administered by a low-cost professional and that the costs associated with caring for individuals with substance abuse problems tend to be high.
The Primary Care Behavioral Health model has also demonstrated improved outcomes for depression and general mental health functioning, although studies have tended to be smaller in scale and not randomized. The model has been shown to be highly acceptable to patients who demonstrate significant therapeutic alliance even in the brief episodes of care typical to the model (Corso, et al., 2012) and has shown sustained general clinical improvement over time even among patients with significant initial pathology (Bryan, et al., 2012). In addition the model has been demonstrated to improve the frequency of use of screening tools, reduce referrals from primary care to specialty care and reduce reliance on antidepressant medication use (Serrano, et al., 2011). One study set in a family medicine residency program found that a blended approach using the PCBH model and care management associated with the Collaborative Care model showed the most improvement for depressive symptoms compared to either co-located mental health care or PCBH alone (Landis, et al., 2013).
All these models have demonstrated improvements compared to TAU, perhaps as a result of how poor usual care typically is. The future of research lies in identifying the key aspects of each of the models that are essential ingredients for the improved outcomes and the ideal settings where each model is best developed. A glaring weakness of the literature at present is the lack of implementation analyses assessing how sustainable these models are both from a payer standpoint and from a practice acceptability perspective. These analyses must substitute providers’ general statements of acceptability with true assessments of how well integrated these models become in practices and how they are sustained over years, especially in absence of research conclusively demonstrating enhanced outcomes.
As the demand for integrated care has grown a key challenge has emerged in the area of workforce development. While studies are promising with regard to clinical outcomes for integrated care models little attention has been paid to creating and nurturing a sustainable and scalable workforce to meet the population need. In recent years several programs have emerged to attempt to meet this challenge, although in general the state of workforce development efforts are best described as disparate and relatively isolated from one another. Workforce development has generally occurred via three types of methodology: individual consultation/ technical assistance, academic/ curricular approaches, and learning collaboratives or networks.
Given that academic centers have tended to lag behind on-the-ground innovations in integrated care much of the workforce development has occurred outside of academic centers and much of it has occurred in informal or semi-formal fashion. Clinics with established programs are often asked to provide technical assistance to other clinics or systems looking to replicate their work. In recent years some of this technical assistance has been formalized either by individual consultants or by programs that have taken this work on. One such example is the Center of Excellence for Integration, a program of the North Carolina Foundation for Advanced Health Programs. The Center of Excellence is an example of a grant funded, health-oriented, non-profit foundation that provides technical assistance to grantees throughout the state of North Carolina seeking to achieve integration of physical and mental health care. The Center’s work takes the form of on-site training of mental health professionals seeking to re-specialize as well as a range of activities to promote integrated care, including service on state task forces and community presentations on integration topics. The Center is funded in large part by the Kate B. Reynolds Charitable Trust (kbr.org) which designates grantees having both interest and capacity to implement integrated care efforts in underserved communities. The centralized nature of the technical assistance allows for some measure of consistency of implementation across sites, although sites can elect to achieve different levels of integration for their particular purposes.
A few academic programs have responded to the need to develop an integrated care workforce. These include degree based programs that have developed integrated care tracks, one degree based program dedicated entirely to training integrated care professionals (Arizona State University Doctor of Behavioral Health), and certificate programs offering online instruction for re-specializing mental health professionals. The Arizona State University program is specifically geared towards master’s level mental health professionals and confers a doctorate in behavioral health, though no licensure based on this doctorate is currently available. Certificate programs include the University of Massachusetts Medical School Center for Integrated Primary Care and the University of Michigan Web Based Certificate In Integrated Behavioral Health and Primary Care. Fairleigh Dickinson University has held a certificate in integrated primary care since 2011 which is now being transitioned to continuing education modules for the National Register. The University of Washington has also developed training resources through its center called Advancing Integrated Mental Health Solutions (AIMS) for institutions interested in implementing the collaborative care model. These certificate programs have filled a vital need for providing a core curriculum, usually of around 40 contact hours, that provide basic orientation and disseminate best practices for mental health professionals and/or care managers and allied health professionals in working towards integration.
Learning collaboratives are another strategy that have been used to attempt to scale training methodologies. One example of such a collaborative is hosted out of a nonprofit called The Health Federation of Philadelphia (healthfederation.org). The Health Federation works with a variety of Philadelphia-based community health centers and public health clinics interested in integration by providing technical assistance for implementation of integrated care based on the Primary Care Behavioral Health model. The Health Federation hosts monthly group learning sessions for all of the mental health professionals operating as Behavioral Health Consultants in their respective organizations in addition to some education sessions for primary care providers and administrators. A key innovation of this learning collaborative has been the implementation of a simulated patient exercise where the core competencies of the Behavioral Health Consultants are tested annually for the dual purposes of measuring the impact of the learning collaborative as well as providing feedback to the Behavioral Health Consultants with respect to their growth. Learning collaboratives are often amenable to group training, this helps expand new professional identities and makes delivery of education scalable.
One strategy that has grown significantly in the last decade is expansion of postdoctoral training opportunities. The number of such postdoctoral opportunities has grown substantially particularly through systems that have embraced integrated care such as community health centers and Veterans Administration centers. Although there is no standard training methodology across all of these postdoctoral opportunities, they typically provide supervised experience in a primary care environment along with some program development activities given that many of these sites are still early in the development of their integrated care practices.
In addition to the curricula outlined above interested mental health professionals now also have the opportunity for self-study using one of several texts that have been written in the last decade. These include texts written related to the contents of typical integrated care visits (Hunter, et al., 2009), the nuts and bolts of program development (Serrano, 2014; Robinson, et al., 2006) and ecological issues (Talen, et al., 2015) related to building integrated care programs.
There is still a substantial amount of work needed in the area of workforce development if integrated care is to become the standard of care. To accomplish this traditional academic centers and accrediting bodies for respective professional groups will need to make significant adjustments to allow for the flexibility needed to create and implement integrated care curricula and training experiences.
As practices have developed integrated care programs the need for change in federal, state and local regulations and policies has become apparent. Private and public policy changes spurred by integrated care have ranged from new provider documentation guidelines to larger experiments in provider reimbursement. The current landscape is ripe with innovation and experimentation. Two such exemplars include the states of Oregon and Colorado.
The state of Oregon has piloted a restructuring of its Medicaid system whereby physical and mental health providers are organized together into what are called Coordinated Care Organizations which integrate systems in an effort to reduce costs and improve patient outcomes. An aspect of this structure is a pay for performance methodology that makes payment dependent on meeting Coordinated Care Organization quality goals. In its 2014 year end report the state reported that 13 of 16 Coordinated Care Organizations received 100% of the eligible quality payments (Oregon Health Authority, 2015). In addition, a law passed in July 2015 (S.B. 832) codified definitions of behavioral health clinicians working in integrated settings and paved the way for reimbursement of these clinicians in both integrated primary care and mental health settings. Efforts such as those in the state of Oregon have been years in the making and have required significant modifications at the provider, payer and regulatory levels.
The state of Colorado has also been at the forefront of experiments in integrated care with a number of health systems implementing integrated care programs. In June 2015 the Colorado State Innovation Model project announced that six major payers and the state Medicaid office had come to an agreement on a four-year plan to implement delivery and payment reforms aimed at integrating physical and mental health care (State Innovation Model, 2015). While the specific reforms are yet to be released the end goals include reducing costs while streamlining accessibility of mental health care in primary care settings. Similar to the innovation in the state of Oregon, Colorado’s Medicaid office has been piloting quality incentivized contracts in coordinated care organizations since 2011.
While statewide and federal policy changes garner a great deal of attention and are essential for scaling integrated care, considerable effort has also occurred on the local level to enable integrated care models to succeed. The Health Federation of Philadelphia has worked with the local Medicaid managed care organization to create new billing codes for Behavioral Health Consultants and streamline primary care documentation. The Wisconsin Psychological Association has worked with state regulatory boards to synchronize state and federal HIPAA regulations and to remove arbitrary barriers between physical and mental health record keeping, thereby facilitating provider communication. In aggregate, these changes represent substantial progress towards creating an ecology where integrated care can thrive. Other examples of policy innovations can be found here: http://www.integration.samhsa.gov/about-us/esolutions-newsletter/e-solutions-june-2015.
Among professionals who have worked in integrated care for the last 15 years a common refrain is that it is no longer a question whether integrated care is here to stay but rather how quickly it will become the standard of care and in what form. The basic outlines of what integrated care can look like have been delineated and what remains is a challenge of creating an ecology that will support various forms of integrated care. This includes challenges related to persistent barriers to payment for integrated services, the development of business models in nonintegrated settings, training and credentialing of integrated care practitioners, and the ever-present challenge to continue to develop research that addresses both clinical, cost-effectiveness and program sustainability components. In addition, the very human challenge of guild issues and identification of often-underappreciated champions complicate progress towards true integration.
Payment reform remains essential, and will have to include solutions to the problems associated with prior authorization requirements and billing issues that prohibit organizations from billing from “same-day billing” for multiple services for the same diagnosis. Clinical practice standards must change to allow alternatives to the traditional comprehensive mental health evaluation. It is likely that payers will find it more advantageous to eliminate these barriers while controlling their cost ceilings through capitated and bundled payment strategies that likely will also include quality incentives for providers or provider groups. True to the structure of the US health system these arrangements are likely to vary significantly from locality to locality. In addition payers will have to figure out how to reimburse for activity that does not involve office visits such as written consultations from a psychiatrist or a curbside consult from a Behavioral Health Consultant.
...for psychologists, integrated care represents a unique opportunity to provide leadership in areas that are truly distinctive to the profession...
While much of the current growth in integrated care has occurred in integrated settings such as the Veterans Administration and community health centers, there are many settings where the physical and mental health providers do not belong to the same organization. Business models that are financially sustainable for physical and mental health providers in these arrangements are complex and will be one of the emerging areas in integrated care. There are some parallels in the current health system such as the way that some hospitalists or emergency medicine providers are paid by hospitals via private group contracts, but these arrangements are in their nascent stages in the world of integrated health care.
Even if payment issues are resolved it is arguable that the largest issue of all facing the integration movement is that of training a qualified workforce. There are no current national standards that define the core competencies of a care manager, a consulting psychiatrist or a Behavioral Health Consultant. A credentialing process that reflects training and vetting of qualified individuals across professional categories is an essential element of building a professional workforce and will become increasingly essential to health care payors demanding both quality and accountability. Guild issues are likely to be a key issue here as these functions in integrated care environments can be performed by a variety of professionals. For example, a psychiatric nurse practitioner or a prescribing psychologist can serve the function of a consulting psychiatrist and both a licensed clinical social worker and a clinical psychologist can serve the function of a behavioral health consultant. How to adopt consistent interprofessional credentialing, and the cost of doing so, remains a significant obstacle. This of course presupposes consistency between guilds in training methodologies, ideally resulting from collaborative efforts to avoid disparate core competencies and which ideally also should be a collaborative effort between guild groups to avoid disparate core competencies and therefore a disparate product to provide the public and payers of healthcare.
The future of integrated care also depends on the continued development of research demonstrating not only the clinical efficacy of integrated care models but detailing the costs associated with this form of intervention, the potential for cost offset and the elements that promote true sustainability. Sustainability in particular is a weakness in the literature given that research projects are often performed in discrete episodes and are more often oriented towards immediate outcomes. Since integrated care fundamentally represents health system reform, a longitudinal approach to studying program development efficacy and sustainability is arguably a truer method of determining best practices than a two-year trial. What will ultimately win the day in integrated care is what sticks over a ten-year period. This means that factors such as workforce satisfaction, ease of implementation and sustainability of protocols as well as costs and workforce retention are aspects that need further attention.
Finally, for psychologists integrated care represents a unique opportunity to provide leadership in areas that are truly distinctive to the profession, including leading program development and evaluation projects, creating curricula, providing clinical supervision and bridging the gap between our colleagues in medicine and mental health specialists. However, integrated care also poses a challenge to psychologists in that it requires a uniquely new professional identity that is in many ways distinct from our traditional professional identity. Those who have worked in integrated care settings in the last 15 years can attest to the ways in which the medical world has become more of a professional home to psychology than the world of specialty mental health care. This is something to be embraced but will pose a challenge as we seek to continue to remain connected to specialists in mental health as well as our colleagues in medicine. Given the current zeitgeist, the future is bright for psychologists interested in integrated care.
Dr. Serrano is an Associate Director at the Center of Excellence in North Carolina where he provides technical assistance to health organizations across the state to improve or initiate integration projects. He has devoted his entire career to working with federally qualified health centers (FQHC), starting integrated care programs and consulting with clinics in underserved settings to assist with implementation of primary care behavioral health (PCBH) programs. Dr. Serrano's research interests include program development evaluations and outcome studies related to PCBH, particularly in underserved settings. In 2014, Dr. Serrano edited an e-book titled, "The Implementer's Guide To Primary Care Behavioral Health," a practice management handbook. One of Dr. Serrano's most outstanding contributions to the field of psychology has been his passion to teach and train the future PCBH workforce. In 14 years of practice he has trained more than 100 students and professionals in the practice of Behavioral Health Consultation in primary care.
Barbosa, C., Cowell, A. J., Landwehr, J., Dowd, W., & Bray, J. W. (2015). Cost of Screening, Brief Intervention, and Referral to Treatment in Health Care Settings. Journal of Substance Abuse Treatment. pii: S0740-5472(15)00141-5. doi: 10.1016/j.jsat.2015.06.005. [Epub ahead of print]
Bogner, H. R., Morales, K. H., de Vries, H. F., & Cappola, A. R. (2012). Integrated management of type 2 diabetes mellitus and depression treatment to improve medication adherence: a randomized controlled trial. The Annals of Family Medicine, 10(1), 15–22.
Bryan, C. J., Corso, M. L., Corso, K. A., Morrow, C. E., Kanzler, K. E., & Ray-Sannerud, B. (2012). Severity of mental health impairment and trajectories of improvement in an integrated primary care clinic. Journal of Consulting and Clinical Psychology, 80(3), 396–403.
Corso, K. A., Bryan, C. J., Corso, M. L., Kanzler, K. E., Houghton, D. C., Ray-Sannerud, B., & Morrow, C. E. (2012). Therapeutic alliance and treatment outcome in the primary care behavioral health model. Families, Systems, & Health, 30(2), 87–100.
Estee, S., Wickizer, T., He, L., Shah, M. F., & Mancuso, D. (2010). Evaluation of the washington state screening, brief intervention, and referral to treatment project: cost outcomes for Medicaid patients screened in hospital emergency departments, 48(1), 18–24.
Huffman, J. C., Mastromauro, C. A., Sowden, G., Fricchione, G. L., Healy, B. C., & Januzzi, J. L. (2011). Impact of a depression care management program for hospitalized cardiac patients. Circulation: Cardiovascular Quality and Outcomes, 4(2), 198–205.
Hunter, C.L., Goodie, J.L., & Oordt, M.S. (2009). Integrated behavioral health in primary care: Step-by-step guidance for assessment and intervention. Washington D.C.: APA.
Katon, W., Russo, J., Lin, E. H. B., Schmittdiel, J., Ciechanowski, P., Ludman, E., et al. (2012). Cost-effectiveness of a multicondition collaborative care intervention: a randomized controlled trial. Archives of General Psychiatry, 69(5), 506–514.
Landis, S. E., Barrett, M., & Galvin, S. L. (2013). Effects of different models of integrated collaborative care in a family medicine residency program. Families, Systems, & Health, 31(3), 264–273.
Oregon Health Authority, (2015). Oregon’s health system transformation: 2014 Final Report. Retrieved from: http://www.oregon.gov/oha/Metrics/Documents/2014%20Final%20Report%20-%20June%202015.pdf
Robinson P.J., & Reiter J.T. (2007). Behavioral Consultation and Primary Care: A Guide to Integrating Services. New York, NY: Springer.
Serrano, N. (Ed.) (2014). The Implementer's Guide To Primary Care Behavioral Health. Retrieved from: https:// itunes.apple.com/us/book/implementers-guide-to-primary/id833906873?mt=11
Serrano, N., & Monden, K. (2011). The Effect of Behavioral Health Consultation on the Care of Depression by Primary Care Clinicians. Wisconsin Medical Journal, 110(3), 113–118.
State Innovation Model. (2015). Colorado health plans make expanded commitments to integrated care and state innovation model. Retrieved from: http://www.prnewswire.com/news-releases/colorado-health-plans-make-expanded-commitments-to-integrated-care-and-state-innovation-model-300103461.html
Talen, M.R., & Valeras, A.B. (Eds.) (2015). Integrated behavioral health in primary care: Evaluating the evidence, identifying the essentials. New York, NY: Springer.