Group Health Cooperative

Group Health Cooperative (GHC) is a large nonprofit health care system that provides both medical coverage and care in Washington State and Northern Idaho, with approximately 568,000 enrollees. Overall, a staff model is used in more densely populated areas with deeper penetration, while network arrangements are used in less dense areas. The staff model serves about 70 percent of the members. GHC is organized as a community of businesses within the integrated health system with a shared purpose of providing high quality and affordable health care. The organization is governed by an 11 member board of trustees, all of whom are GHC members elected by other members.

Within GHC, Behavioral Health Services (BHS) have tended to run with mixed staff and network models even in dense areas because of the seasonal rhythm to referrals, e.g. Seasonal Affective Disorder. BHS has been involved in a transformational process over the last two decades, responding to the problems of improving access to behavioral health care and improving quality of care, both behavioral and medical. In the early years, throughout the country, behavioral health care was essentially a cottage industry. The advent of managed behavioral care changed standard operating procedures within BHS over time, knitting services together to form a system, and ultimately a business. This transformational process has transpired in several phases and is ongoing.

Integrated care was launched to improve access and quality of care within an organization with a fundamental set of organizing principles committed to systematic care. The fact that BHS was already embedded in a medical care organization was seen as an advantage. Integration was also a response to the threat of carve outs, which had been significantly successful in gaining market share. Historically, carve outs, by definition, tended to reify behavioral health specialty as separate from the population-based care perspective. An over-focus on such a division of labor restricted access, particularly at the point of contact most frequented by people with behavioral health issues, which is primary care.

BHS also had the advantage of being part of a system that has been seminal in integrated care research. The primary investigators of the research also functioned as clinicians in medical and behavioral health. In theory, BHS would have been best placed to implement what was learned from the research. GHC's Center for Health Studies has also investigated effectiveness of treatments in naturalistic settings by embedding intervention in GHC patient services. But the real world is more complicated than even is found in effectiveness studies.

In the early days, preparing the organization for the idea of integrated care required a considerable amount of raising consciousness with regard to mental illnesses. The concept of epidemiological intelligence, influenced by research in the UK, gradually led to the understanding that a population perspective for behavioral health is legitimate and useful. The vast majority of people with mental illness are actually seen in primary care. Also during this time, the managed care environment in the US generated the National Committee for Quality Assurance (NCQA), which included depression care medication management as a quality indicator. This helped spur support for organizing a “roadmap for depression”, which used electronic charting to improve depression care follow through. GHC's improvements have held over time, with 75^th to 90^th percentile marks for the depression Healthcare Effectiveness Data and Information Set (HEDIS) indicators.

BHS was involved in a second initiative as well, this one without formal department sponsorship. BHS established a business relationship with primary care to co-locate clinical staff in area medical centers on a part time basis to be available for general consultation. Specifically, a psychotherapist would spend 20 percent time in a medical center for 30 minute consultations with patients with psychiatric problems that were unlikely to be referred for specialty care. The purpose of the initiative was to improve access to behavioral health care and take advantage of efficiencies for patient convenience and to intervene at the initial site of concern, primary care. Within a utilization corridor, if behavioral health penetration, base of utilization, increased by 10 percent, primary care would reimburse BHS with a per member per month fee. If penetration did not increase, or declined, BHS would reimburse primary care.

The major effort for the primary care general consulting program focused on training behavioral health clinicians to function more like primary care providers; the 15 minute primary care clinic visit versus the 50 minute hour behavioral therapist visit. The BHS therapists involved in the initiative reported enjoying the new environment, and the program was popular. Given that primary care general consultation visit was usually a 30 minute visit, the BHS therapists were making themselves available for more patients within a work day. This was part of the basis for the informal reimbursement agreement between primary care and BHS.

In fact, penetration did increase by more than the required 10 percent in the Seattle area, but the late 1990s was a financially challenging time for the organization in general, and primary care was unable to afford the within-company reimbursement. So, even though the BHS initiative was available within a staff model HMO and single payer, finances still brought the initiative down.

Overall, these experiences taught BHS that, in order to compete with carve out competitors, they would need to take on business properties such as knowing the competition, understanding cost structures, and having solid assessments of good performance. BHS was trying to balance collaboration and consultation on the one hand and performing to industry specifications as represented by carve outs and HEDIS on the other. It was a classic case of needing to focus on what are deemed important business indicators as represented by the carve outs and HEDIS, which was a limiting factor in allowing the necessary increased resources to meet the integration opportunity.

From the 1990's, BHS's focus increasingly turned to running a business model and hitting the quality indicators. Depression care, a la HEDIS, was an area that was doing well, but the primary care general consultation program was discontinued and primary care and behavioral care returned to traditional models.

The next growth phase for integrating care came with the implementation of a new electronic medical system which included both medical and behavioral health information. Considerable effort was spent on designing the system, and there were adaptive issues around how to balance sharing information between providers with confidentiality requirements. A split clinical note was developed that had one section for the behavioral clinician to record confidential patient information. A second section with assessments and treatment plans which could be shared with medical providers when there is a clinical need to access such information.

Even with the upfront time commitment to developing the EHR, though, the launching was met with mixed success with the medical staff. There was a conflict of cultures over how the therapists documented cases and what the physicians felt they needed in order to help and follow through with patients under treatment. There was also still an unmet need of improved integration that could be accomplished by sharing some information with nurses, pharmacists, and social workers. The EHR was changed to allow access to these other disciplines. A warning system was installed that required the user to input a log-in password and a reason for accessing the record for each and every encounter. This was viewed as over-burdening by the medical staff as well, and future changes will be coming.

In the current business environment, BHS has been seeing a synergy developing between integrated care processes and business indicators. For example, the National Business Coalition for Health (NBCH), and the affiliate group, the Puget Sound Health Alliance, have been monitoring the HEDIS indicators for ADHD, alcohol, and depression. Good systematic tracking and follow through by health organizations is required to achieve high marks on these indicators.

Further, the Puget Sound Health Alliance has developed an accreditation process, EVALUE8, which is a set of questions, like accreditation standards with measurements somewhat like HEDIS, including those that are pertinent to integrated care. NBCH is looking for evidence of processes such as case identification (PHQ-9 for depression or AUDIT for alcohol), conventional and non-face-to-face outreach efforts (telephone and internet), and the care organization's ability to report follow through with the processes. If EVALUE8 is successfully implemented, it has the potential to demystify integrated care and send a clear signal about what is involved in the follow through of clinical processes.

GHC is also investigating the Toyota system LEAN which focuses on processes and uses outcomes to perfect the business's clinical functions. GHC is very committed to using LEAN to provide clinical care, including integrated care.

BHS has also been moved into the primary care business structure within the GHC organization, which places them even more centrally to follow through with integrated care. They are continuing to pursue NCQA accreditation with the QI 11 standards and guidelines focused on continuity and coordination of care between medical and behavioral health services. Attention is being placed on information exchange, psychiatric involvement in formulary choices, and adherence monitoring. General consultation is available in the form of Mind Phone, a psychiatry telephone consultation line. Psychiatrists divvy the work time, manning the phone during the work week to assure someone is always available to all GHC clinics for questions. There is also a focus on prevention and monitoring of medical risks for patients using psychopharmacology, for example, elderly patients on tricyclics for sleep problems when they face other increased health risks.

RESPECT-Depression Dissemination

RESPECT-D was designed not just to test an integration model, but also the ability of a model to be disseminated across organizations. The RESPECT-D research team conducted an extensive qualitative investigation into the factors contributing to successful implementation and dissemination, or the barriers to implementation, after the research trial concluded. ²¹¹ Two of the five HCOs involved in the trial, both of them medical groups, continued with the TCM and expanded it to all clinics. The following lists the major lessons from the article.

TCM strategies. The PHQ-9 was widely seen as the most useful of the TCM components. Many physicians continued to use it for confirming diagnoses and monitoring patients, even after all other program components were discontinued.

Psychiatric oversight of care managers was widely valued by clinicians, care managers, and the mental health specialists who appreciated the ability to provide expanded support to a larger number of patients.

The large majority of communications involved medication management and psychiatric comorbidities.

There was a nearly universal failure of the clinicians to distinguish between self-management support and general patient education. The care managers, who were responsible for providing the self-management support, were more likely to understand the difference and view self-management support as an important component of care.

Care managers were also valued by clinicians, although this opinion was tempered by the time required for communication and the cost of additional staff. The locations and way care managers were used changed post-trial for continuing HCOs. Care managers tended to be located onsite, and there was wider variation on patient characteristics PCPs relied on to select which patients they felt would benefit from referral to care management.

General clinician perceptions. Changing a practice is very difficult and not worth the effort unless it makes a big difference; change that only improves care for a single disease is often not seen as efficient.

While care managers were valued, physicians felt burdened by the time spent in communication with care managers, or attending to care management forms, even if only “a few minutes here, a few minutes there.”

Most physicians were loath to link services to a health plan, providing improved care to only those patients with the proper coverage.

Organizational characteristics associated with sustaining and disseminating TCM. HCOs that successfully disseminated the TCM to all clinics had “a mission and vision of improved care that was widely shared among leadership of the organizations and clinicians at the practice level.” This commitment extended beyond depression care to include chronic care in general.

The HCOs were committed to a clearly defined and widely-understood institutional change strategy in place before the trial began. The HCOs had a history and culture of improvement change, including systematic change.

Leadership was clearly associated with successfully sustaining and spreading the program.

The ability to rationalize the cost of the program was also key to implementing and spreading the program. The rationalization may be clinical—“it's good patient care”—rather than directly economic.

Implementation was easier the more the clinics followed staff models and organization provided an integrated system of care.

Adoption of the TCM in the two medical group HCOs was part of a larger vision and more comprehensive initiative to improve chronic care.

Organizational characteristics associated with decision not to disseminate TCM. Many PCPs who participated in the trial through two health plans disliked providing improved care to only those patients with the proper coverage. The PCPs did not wish to limit improved care to only a select group of patients. They also saw a loss in efficiency when administrative practices are applicable to only a subset of patients.

The loss of leadership is just as strongly associated with the inability to sustain the program. Two of the three HCOs not continuing had experienced loss of visionary leadership during the trial.

Staff model relationships between the clinics and HCOs are not enough in the face of the lack of an economic model; nor is a staff model a guarantee that a noneconomic justification for the program will be successfully adopted.

Health plans had the most difficult time implementing, sustaining, and spreading the TCM. Neither health plan participating in the trial had more than indirect influence through reimbursement policies over the participating clinics, nor were they able to change their reimbursement policies within the context of the TCM.

Eastern Band of Cherokee Nation

The Eastern Band of Cherokee Nation (also known as the Eastern Band of Cherokee Indians, or EBCI) Health Service is a largely rural network of health services. Any person identified as a member of a federally recognized tribe is eligible for services. Approximately 10,000 of the 14,000 EBCI members are users of the tribes' health care system, which is governed by many tribal and federal government rules. Under self-governance, the EBCI runs one 16-bed hospital with one onsite and one offsite outpatient clinic and five tribal outpatient clinics offering primary health care services. Funding for the system is from four primary sources: the Indian Health Service (IHS), tribal funds, reimbursements from other health payers, and grants. Tribal funding, particularly from Indian gaming, has become a significant proportion of total funding; it has been demonstrated nationally that federal funding through the IHS is insufficient and lower than that provided for prisoner health care. Patients who require specialized services or tertiary care not available within the network are referred out to receive contracted services from providers in surrounding areas. From the patient perspective, EBCI functions as a single payer health system. EBCI will bill any eligible third party payer, such as private insurance, Medicare, and Medicaid, conserving its funds as a payer of last resort.

The EBCI integrated care program targets a specific population, the Indian members, rather than a clinical problem, such as depression. The program began as a bottom-up initiative introduced by mental health staff. A child psychologist offered to locate part-time in primary care clinics and school health offices to bring the services to where the patient/clients are. Access to the new venues was created by building relationships with the primary care providers. The initiation and on-going development of integrated care was done with the awareness of the health system management. They “gave their blessings to what the folks in the field were working out,” according to one psychologist. The idea of making services available to the patient in their place of choosing was a major contributing factor to the development of the program.

Currently, different locations are scattered across levels 1 through 4 of Doherty et al's. 5 Levels of Systemic Collaboration model. ⁵⁶ The most fully integrated services are available for the pediatric population, up to age 22, but integration has been gaining traction in the adult population as well. Mental health specialists are co-located part time at rural medical clinics for both adults and pediatrics. Some locations are “just borrowing office space.” Other locations make use of the possibility of informal consultation, and one provider may pull the other team member into a clinic visit, be it the mental health specialist or the PCP, for a quick joint consultation with the patient. The level of integration development depends on the state of the relationship between the providers.

Mental health providers co-located in a clinic make themselves available to consult with the PCPs on behavioral health issues, including joining in a patient visit. Similarly, when providing behavioral health services to a client, the mental health provider may ask the physician to join the client visit to address a particular medical concern. The mental health providers also monitor patient progress, including medications. All patients on medications are required to remain in contact and visit a therapist at least annually. Mental health providers that do not have a terminal degree are supervised by psychologists and psychiatrists.

Integration programs that are problem focused are also being developed. As primary care providers have success with referral to co-located mental health providers, they are more open to implementing behavioral health services for patient self management. Integrated care is provided in a pain management clinic and with a new diabetes care management program that includes integrated depression management as a comorbid condition. There is also a new teen model being developed for common teen concerns that includes relaxation and cognitive behavioral training. Substance abuse treatment programs are also linked to primary care clinics to improve patient followup. Psychiatrists have also been working with the health system formulary for appropriate psychopharmacotherapy choices.

The EHR makes available to all providers the full medical and mental health life history. The system uses notes with signoff requirements to facilitate communication. Stepped levels of security exist for medical versus mental health records, allowing the mental health provider to set access for primary care providers for individual patients if it is deemed necessary to the patient's care.

In November 2007 a partnership of EBCI, Western Carolina University, and the Jackson County Department of Public Health was awarded a grant of $3.6 million to develop and extend a broadband telehealth network. The EBCI will use the new infrastructure to increase access to mental health services through telepsychiatry. The psychiatrist on staff at one of the tribal outpatient clinics will be able to provide services to more remote locations. There is also anticipation that telepsychiatry may function as a culturally sensitive tool for mental health care for some members who are more remotely located.

Staff with EBCI considered integrated care and its holistic view to be a natural fit with the culture of the organization and tribal governance and clientele. Not surprisingly, patients are often less likely to distinguish between mental and medical health than health care systems have been historically.

Growth of the program has been allowed to remain fairly organic and bottom-up in orientation. Rather than imposing change, providers have the opportunity to observe the benefits and positive outcomes of co-location and access to the tools and services mental health specialists provide and ask for the services to be made available in their location as well. There was also some demand for integrated services created by the PCPs, recognizing that the outcomes for substance abuse treatment were not acceptable and wanting improvements.

Coordination and openness to collaborate required adjustments by both medical and mental health providers. Medical personnel, including nursing staff at the clinics who were most familiar with referring a patient out, needed to learn the potential benefits of remaining in the treatment program with integrated care. The mental health specialists also had to adjust to the primary clinic setting, where the 50 minute hour might include interruptions for a quick consultation on a different case, just as PCPs often are. Physicians often started the collaborative process with a particular condition, such as depression or ADHD, and expanded out as they gained familiarity.

While EBCI is not a system with a wealth of resources, financing was not the main barrier to implementing integrated care. Staff needed to provide integrated services are salaried and paid for by EBCI, so historically the focus has not been on billable services. EBCI has creatively used grants where possible to cover investment in new technology or start-up costs of new initiatives. Billing of third party payers remains a focus to improve revenue wherever possible. The largest barrier to integrated care has been relationships and time, but perhaps the word “barrier” is not the proper word when viewed from the organic perspective applied by this program.

Tennessee Cherokee Health

The Tennessee's Cherokee Health System's (CHS) integrated care focus began in 1978 as mental health outreach from a community mental health agency, rooted in a public health model. The mental health outreach targeted primary care for the simple reason that primary care was where the patients were located and it allowed the patients to move past issues of stigma. In 1984 the agency recruited a primary care physician, borrowed money to build a clinic, and opened its first integrated practice. In addition to specialty mental health care and dental services, CHS currently operates 14 integrated clinics in 11 East Tennessee counties, providing an array of comprehensive primary care and mental health programs for adults and children. There are over 50,000 clients served annually by CHS.

CHS's integrated care model developed over time more by virtue of experience than by application of theory. The treatment model which has evolved features a Behavioral Health Consultant (BHC) embedded in the primary care team and providing care different from typical mental health models. Psychiatric consultation is also available to the primary care team. Clinical practices evolved as clinicians found that behavioral health services were helpful to people with chronic diseases as well as for all the psychiatric disorders that present in primary care.

Cultural changes were experienced by both behavioral and medical care providers. It was initially difficult for new mental health staff to understand the dominant primary care culture of the integrated clinics. Most mental health providers weren't—and aren't—trained in primary care settings. The reward to practicing in the primary care setting is expanded access to clients. Similarly, PCPs were unfamiliar with mental health care processes and potential benefits. Currently PCPs are seeking out CHS for employment, specifically for the benefit of working in the integrated environment. The PCPs don't have to “sell” a referral to specialty mental health care to patients, and they don't have to worry about accessing help for difficult patients. PCPs are quoted as saying “I know that if I ask that question, the patient will dissolve into tears and I just don't have the time. Here I have an easy hand-off.”

Integrated care is delivered by teams with shared decisionmaking among the team members. No one team member is assigned primary responsibility for aspects of patient care such as medication adherence monitoring. All team members have access to the treatment plan and support it. The ideal clinic build-out (not all clinics are able to accommodate this ideal) uses a pod structure with the BHC centrally located in the midst of the exam rooms. This facilitates the co-management of care and constant team communication. A patient may be handed back and forth between mental health and medical providers within one clinic visit or meet jointly with both providers. The BHC, usually a licensed clinical psychologist or licensed social worker, is a generalist, just as PCPs are generalists. On a given day a BHC may be involved in not only interventions for mental health concerns but also health education or lifestyle change; whatever supports the treatment plan.

Teams meet for weekly team meetings in all clinics and all clinicians attend. Considerable training and cross-consultation occur in the meetings. The team meetings are the mechanism for shaping culture and building clinical models. Clinical models are not handed down through administrative processes.

CHS functions as a hierarchically flat organization. Providers, both mental health and medical, are spread over large geographic areas with variations in practice sizes and each clinical team has significant autonomy. An EHR system is used; this system steers data collection and helps shape the clinical model to some extent. With the EHR, CHS providers use simple standardized screenings with a few red flag questions.

Patient records on the EHR are available to all team members, including treatment plans. Each patient signs a consent form that acknowledges treatment is provided by a multidisciplinary team and all providers will have access to patient records. CHS expends extensive training around appropriate documentation—what should and should not be recorded in a clinical record—because it represents one of the major changes from the typical mental health model for the BHCs. Information needs to be in the form required for primary care services; brief, succinct, on task. Unnecessary personal information about patients should not be included. The BHCs are trained to think of the service they provide as a primary care service, with specialist mental health services available by referral to specialty mental health providers who are also available within Cherokee.

The payer distribution of CHS patients is 41 percent TennCare/Medicaid, 23 percent self pay, 19 percent commercial insurance, 13 percent Medicare, and 4 percent supported by other contracts. When TennCare was implemented, the state funds for community mental health sliding fees were diverted to TennCare. Thus, state funds to support uninsured patients are limited. CHS uses the strategy of negotiating for global funding streams—capitation, percent of premium, case rates, anything other than fee for service—whenever possible because that allows providers to focus on care, not billable units.

CHS data shows that patients enrolled in Cherokee's behaviorally enhanced health care home had lower utilization of specialty mental health services and subsequent primary care visits. Dr. Dennis Freeman, Cherokee's CEO, sees evidence that integrated care is beginning to be viewed by payers as cost-effective, and the plans are beginning to be willing to pay for it. Blue Cross Blue Shield of Tennessee compared service utilization data for CHS patients compared with patients of other providers in the region. They found CHS patients had higher PCP utilization, a favorable finding given their emphasis on patients using their healthcare home. They also found lower:

• Overall costs per patient
• Specialist utilization
• ER utilization
• Hospital admissions

Once a financial structure is in place, the real work comes in finding the right behaviorist with the right personality, skill set, and work style, to blend into the patient care environment. CHS leadership's major focus for consultation work with organizations wishing to add integration often involves recruiting and mentoring the behavioral health staff's new way of practice.

There is evidence the payers are catching on. There is a coalition of governmental bodies (including CMS, SAMHSA, and HRSA) that are focused on financing integrated care. Recently they issued a report examining the reimbursement of mental health services in primary care settings. There are growing numbers of workshops and conferences on integrated care. Managed care organizations and state Medicaid programs are moving away from a carve-out environment and into a carve-in environment. Departments of mental health from other states are interested and are contacting CHS to learn more about how they provide integrated care.

Since CHS is a comprehensive integrated care program, treatment for alcohol related concerns is a standard practice. CHS practitioners use the first two questions of the CAGE questionnaire as red flag screens. The PCP is also likely to ask a few additional questions of patients. The SAMHSA Treatment Improvement Protocols series manual for interventions in primary care is a valuable resource. There is good evidence that counseling by a physician does have an effect on subsequent drinking behavior. If a warning isn't enough, the BHC is likely to be pulled in for a more thorough assessment and perhaps referrals to treatment programs within CHS. Referrals to detox or inpatient units outside of CHS are also accessed. The care team will track patient followup as well.

CHS does not use many standardized processes across the integrated clinic locations. It is believed that adding standardized processes would be difficult, since they would be counter-cultural to the autonomy and flat organizational structure currently in place. In fact, there is some skepticism towards the specialist behavioral health notion that one sees in the literature, such as depression care managers. Dr. Freeman feels the generalist approach is necessary. The PCP has to deal with everyone that walks in the door, and the BHC should be able to as well.

Dr. Freeman believes integrated care is the future of primary care and community mental health. Community mental health facilities are struggling in every state he visits. “With all we know about how important self-management is for health status and how behavioral concerns factor in, the integrated model is the most logical clinical model for primary care. Add to that the data about the poor health status of many patients in community mental health, and a blending of the two sectors seems advisable.”

CHS views their organization and employees as missionaries. CHS believes integrated care is a better way to deliver primary care. Taking the model on the road is part of the strategic plan that has been a living document for a couple of decades. There is no wrong door to mental health. We have hit the tipping point. There has been a real shift at the organizational level, at the federal level, where people go for mental health services, and growing acceptance of behavioral health care as part of health care teams. The future of primary care is the behaviorally enhanced health care home.

Washtenaw County Health Organization

The Washtenaw County Health Organization (WCHO) is a collaboration between the Washtenaw County and the University of Michigan Health System to provide health care and medical homes for Medicaid and indigent consumers of Washtenaw County, Michigan. WCHO serves 24,000 Medicaid (18,000 dually eligible) and 2,000 SPMI (80 percent Medicaid) patients. While WCHO provides what has been referred to in this report as both forward and backward integration services, this case study will focus on forward integration.

WCHO integration efforts took off in 2000 with the signing of state legislation enabling the founding and funding of WCHO as a new governmental entity. The new organization addressed the tendency of organizations to cost-shift indigent consumers to other organizations or facilities by creating the ability to partner among them. Cost savings were also expected from less fragmented care afforded through service coordination.

WCHO inaugurated its first integrated clinic in 2004. Services are provided through partnerships with local for profit primary health care programs, the Community Support and Treatment Services of Washtenaw County for mental health services, and a variety of other community organizations, primary care clinics, and hospitals, for linkages to an array of comprehensive medical and social services. WCHO specifically targets persons with severe and persistent mental illnesses, substance abuse disorders, and/or developmental disability populations.

There are currently eight sites in various stages of implementation, with at least five fully functional. The sites serve a variety of patient populations: adult, pediatric, teens ages 12 to 21 and their children, African Americans, and indigent populations. Locations range from small neighborhood clinics to a general medicine clinic affiliated with the University of Michigan. Each clinic adapts the core integrated model to fit the local environment. Treatment protocols are selected based on high incidence, comorbid conditions specific to the clinic.

WCHO views integrated care as a single stop shopping place, a medical home where the patient/community members needs are met seamlessly, at highest quality, no matter what the population. The Four Quadrant Clinical Integration Model from the National Council for Community Behavioral Healthcare (described in Chapter 1) was specifically selected to guide the organizations efforts; however, models such as the Five Levels of Collaboration, Wagner's CCM, and Strosahl's integration model, also informed WCHO's integration efforts, as well as published literature for evidence-based best practices.

There are several integrated care components. Mental health clinicians and psychiatrists are co-located onsite for regular consultation and patient visits. Psychiatric consultation may happen curb-side or through more formal channels. Psychiatrists, available one-half day per week and primary care providers may treat patients jointly or through a “ping pong” partnership, passing the patient back and forth for a defined period of time. Case management is brief and in partnership with the primary care providers. Case management includes an array of social service needs in addition to medical/psychiatric and psychosocial support needs and is set at about a 1:35 caseload ratio. Brief psychotherapy is also available onsite, with the ability to refer more complex patients to specialty mental health services, including case management tracking of patient follow through.

WCHO uses a web based EHR that is available to all provider organizations with contracts to provide services to WCHO patients. WCHO has also established a data warehouse to track mental health, substance abuse, and primary care service data and performance outcomes. The nature, cost, and service provider are tracked for services. Patient satisfaction and quality of life measures will be added soon for more complete provider and consumer perspectives than are available with only administrative data. The data warehouse is used to track high utilizing patients, ranking the patient contacts from highest to lowest cost services. Program administrators review the cases with the clinic staff and develop action plans that are presented to the patient. The patient and clinic staff review and adjust the plan. If the action plan is for a systemic condition, recommendations for program changes are made.

The data warehouse also plays an important role in new initiatives for formal standardized processes. Diabetes and chronic obstructive pulmonary disease were identified as high cost, high frequency conditions with potential for more efficient care based on data mining reports, and with good evidence-based practices available for implementation.

Funding for the integrated care program is carried out through a shared funding model, and “intricate web” (see Figure 14). ²¹² Money comes into the community of partners and the community, represented by the WCHO board, then figures out how to pay for services provided. Financial incentives are aligned through risk sharing. All the partners, as director Kathleen Reynolds says, “have some skin in the game.” WCHO has had to be creative and frugal with funding. Only one 2.75 percent increase in state funding has been received since the inception of the program. However, the program has been fully sustainable and not reliant on grants because all partners have come to the table willing to contribute money to the pot. WCHO leadership has been instrumental in fostering this supportive financial collaboration.

Figure

Figure 14. WCHO funding mechanisms.

WCHO uses a simple approach to their collaborative process based on learning organization principles: the rolling start model. Don't wait until everything is in place; begin with the low hanging fruit and build as you go. Taking risks is essential and failure is OK if you learn from it. Use a strengths-based implementation and management process. Build conflict resolution in up front, knowing there will be cultural differences. Follow the decisionmaking plan: determine what is effective, what might help the patient most, and then ask the patient if they want to do it. If the answer is yes, then ask if it is good for the organization. The last question to ask is whether it can be funded. Hire mental health providers who can teach collaboration for onsite clinic positions. Most importantly, follow the philosophy “wait until they ask for it.” Success is more likely when the partner has had time to learn from experience. With a 4-year track record, WCHO is honing in on the necessary model fidelity to track, using a quality improvement structure. WCHO staff knows when a clinic is off model when things begin to break down.

Learning through implementation is a critical element to WCHO's integration efforts. WCHO practices this element through creating and participating in learning communities comprised of the local partners and clinics. WCHO follows again what they view as a simple process: go into new negotiations without the mantle of “expert” and with no agenda beyond helping consumers meet their healthcare needs. Each participant organization and staff person is treated as the expert of their own systems and it is the task of the convened group to figure out what will work best. Start with the leadership and move on to front line staff once leadership is on board. WCHO also supports the national dissemination of integrated care practices through a 50 member learning community involved in similar work, networked through the National Council of Community Health Organizations.

The collaborative partnerships are further strengthened by attention to a CQI approach. Every year the organization looks at what needs to be done differently. WCHO does not add a new service or process unless there is some service or process discontinued in response. This is a new way of doing business within the mental health field and there is a lot of low hanging fruit to go after. Some leadership staff has also been trained in Six Sigma techniques to support the lean quality process. Early numbers for the integrated care program have shown cost-offsets.

This is not to say that WCHO has been immune to cultural change issues. The providers have had to expand their own perspectives and skills and view of biopsychosocial health. When a psychiatrist is only available on site for 4 hours per week, the PCPs have had to learn that the social worker, not traditionally viewed as a peer, was the best consultation source for mental health concerns. Similarly, the mental health providers have had to adjust to a primary care environment with brief visits, quality improvement initiatives, and standardized practices. PCPs have been worried that specialty care psychiatry would “get all the money” through the mental health initiatives. It took time and experience for PCPs to learn that integrated care models allow them to remain central to patient care and can in fact benefit their own mission of providing quality care. “You do it because you become a better diagnostician, a better provider; these are selfish reasons.” “In fact, you don't lose your identity as a provider, but rather enhance it.” However, integrated health will not be reproducible in all offices. Not all physicians will want the expanded scope.

Haight-Ashbury Free Clinics

Haight Ashbury Free Clinics (Haight Ashbury) was founded 40 years ago with the simple goal of providing free medical care to the people gathering in San Francisco for the “Summer of Love.” Rather than ending after the initial identified service need was met, the volunteers, and staff responded to the ever changing and growing need for access to good public health, adding substance abuse treatment and mental health counseling to the primary care services originally provided. Today the Haight Ashbury Free Clinics is one of the largest providers of nonprofit services in San Francisco. Over 200 paid staff and 500 volunteers provide services at over 15 facilities to over 19,000 clients, with the vast majority served by the substance abuse programs.

Haight Ashbury's most recent initiative has been the implementation of an integrated care clinic on Mission Street in the heart of San Francisco. Haight Ashbury's vision of integrated care follows an “any door is the right door” philosophy. The integrated care clinic provides primary care, substance abuse treatment services, mental health services, and intensive case management (which can include referrals to other organizations for assistance with housing, food, clothing, and employment) within a unified team service delivery model. The integrated clinic space incorporates medical exam rooms, group meeting rooms and over 20 individual counseling rooms for mental health and substance abuse services, and 12 social-model detox beds. Staff at Haight Ashbury estimate the new integrated clinic facility will service about 5,000 unique patients, with a considerable portion including patients with alcohol related medical and substance abuse concerns.

The genesis of Haight-Ashbury's integrated care lies with the line staff. Haight Ashbury's organizational culture of advocacy, volunteerism, and looking for ways to best serve the clients that walk through the door was a natural incubator for integrated care. Line staff would notice a particular client's needs and take it upon themselves to talk with other staff and volunteers to determine what would best help the client. Over the years, an informal interdisciplinary consultation network developed. Eventually staff began co-locating where possible to enhance the interdisciplinary approach to care as the benefits became apparent.

A second major contributor to the grassroots growth of integrated care was the preponderance of complex patients in the patient population. Patient complexity comes from many conditions—homelessness, working poor, the physical and mental health sequalae of substance abuse, but he HIV patient population is exemplary. With the HIV epidemic, patients presented with so many health issues, it pushed the line staff to be more attuned to complex patient needs. Since Haight Ashbury had always run as a social model, the staff combined other social support services to help the patients cope with a heavy disease burden and the stigma associated with it. As Haight Ashbury began to be known for the comprehensive approach to complex patients, its reputation drew both providers who wanted to be a part of providing such care, and patients who needed it. Eventually, SAMHSA provided a grant to support the development of integrated care for HIV patients, and many aspects of that program became the prototype for integrated care for the general patient population.

Haight Ashbury is in the thick of instituting processes and systems to support integrated care and grappling with the myriad daily detailed decisions that constitute implementing change. The process currently holding center stage is the charting system. Each of the three services, primary care, mental health, and substance abuse treatment, have their own traditional charting cultures and legal requirements. Combining the three into one comprehensive charting system has involved legal counsel along with cultural and process considerations of the three services. Charting is accomplished with patient records, but the expectation is that an EHR system will available in 2 years if all goes well.

The lack of an EHR system has made the co-location of services in a single facility critical. Communication between staff takes place by email, telephone, or face-to-face meetings. The ability to walk down the hall and talk with a provider from a different service area is crucial. Other systematic forms of communication are also being established and are highly inclusive. For example, weekly team meetings include front desk staff since they are the first point of contact for a patient and thereby necessarily involved in the triage process.

Haight Ashbury's tradition of intensive case management is also a strength being brought to bear for integrated care. Case managers have been primarily focused on the patient population with HIV. Haight Ashbury will need to staff up with more case managers as the therapists hand off to formal systems the informal case management they had been taking responsibility for. Clients meet initially with a case manager and “are literally walked from office to office” by the case manager as they move through the system. The case manager making the initial connections and providing warm hand-offs have been instrumental in patient adherence with treatment plans.

Treatment plans are also expected to be created through fully shared decisionmaking, but this is also still a work in progress. Currently, psychiatry signs off on all treatment plans for all patients with mental health and substance abuse concerns; logistics are still being worked out for medical sign-off.

Even with Haight Ashbury's history, combining services into a single coherent system has had challenges with merging the different service cultures. There are still glitches and adjustments to perceiving how to proceed with one thing or another. Leadership's championing of the home grown strength of integrated care has been essential to settling perceived threats to service territory.

For almost 40 years the organization has functioned on shoe string budgets, focused on the immediate provision of client care with little attention to the thought of creating an organizational and financial model for a sustainable future. Yet, the unsustainable model they did run on—volunteers, grants, unreliable state and local governmental funds—has succeeded in providing uninterrupted services for over 40 years. Partial credit for this lies with the long term staff and volunteers who embodied the institutional memory for the organization.

Over the last 5 years Haight Ashbury leadership has focused on creating a new executive team, strengthening financial controls, restructuring the board of directors, creating a vision for integrated care, and defining for themselves what sustainability is and how it will be achieved.

Haight Ashbury's current funding is approximately 90 percent state and local government general funds, most of it public health community behavioral health service funding. MediCal is a fee-for-service sources of funding. Given California's recent budget crisis, Haight Ashbury has “dodged a bullet” that may have shuttered some or all of their services. While integrated care has not been motivated by a financial model, Haight Ashbury has been focusing on maximizing funding through improved billing, anything that would allow them to take advantage of other reimbursable possibilities.

A major resource for Haight Ashbury is a strong relationship with University of California research faculty. Haight Ashbury has a research arm, the Pharmacology Research Group, which has been conducting clinical trials of medications and therapeutic interventions for addiction treatments since 1990. The research group improves Haight Ashbury's access to grant funds and, by virtue of the protocols under study, can make otherwise prohibitively expensive medications available to clients.

Another subtle support for integrated care found in Haight Ashbury is a long-standing tradition of including complementary and alternative therapies. For example, acupuncture and alternative medicine services have been available since the 1960s and have been used for opiate detox. One volunteer who provides acupuncture services has been with the organization for 30 years. This willingness to cast a wide net to find therapies that work for clients, and the long-lived institutional memory, contributed to creating a fertile environment for integrated care.

The example of Haight Ashbury suggests integrated care is possible in diverse settings. The implementation process at Haight Ashbury has benefited from a prototype program that could be used as a springboard to creating protocols and processes for the larger patient populations. A designated person acts as a central hub for the implementation. Leadership is important, but the person at the hub is the one who carries the comprehensive picture forward when others are focused on the tasks related to their own segment of change. Focus on communication has also been key, relying on organized and persistent point people to assure the communication is reaching all staff effectively. Finally, they are seeking to be efficient at documentation. Each funding source, each grant, adds to the paperwork burden and removes time from client contact. It may feel like golden handcuffs, but the documentation is necessary in order to obtain funding.

Intermountain Healthcare

Intermountain Healthcare (Intermountain) is a nonprofit integrated health care system servicing Utah and southern Idaho. Intermountain has 21 hospital facilities and 200 outpatient clinics at which over 500 staff physicians and 1,000 affiliated physicians provide nearly 50 percent of Utah's health care. Intermountain is committed to the underserved populations and strives to provide the same quality of care across the full rural and urban continuum of Intermountain's facilities.

Intermountain has been providing integrated mental health under a program known as Mental Health Integration (MHI) for a decade. MHI began as a logical extension of a clinical integration structure that organized care by clinical services across the system, rather than by traditional departments, and in which collaborative care was heavily featured. Intermountain's MHI model was developed by a small group of Intermountain clinical leaders. The development was simultaneous with the larger health care environment's introduction of the Wagner chronic care model, the Collaborative Care model out of Washington, and other research initiatives led by integrated care research experts such as Kathryn Rost, and all these sources of research and knowledge informed Intermountain's MHI model development.

One clinic ran a pilot program for MHI in 1998, building on the clinic's previous experience with diabetes and asthma care management practices. Intermountain also leveraged resources that were present at the pilot clinic for MHI. Care managers for other chronic disease conditions were on staff. There were also part time behavioral health staff on site, although they were at the time functioning under a consultation model and ran their services parallel with the primary care services.

The MHI pilot was successful in terms of improved patient functional status and satisfaction, and physician satisfaction and confidence in managing mental health concerns, with neutral cost effects at the clinic and health plan level. With grants from the Robert Wood Johnson Foundation and MacArthur Foundation, Intermountain rolled the MHI program out to seven clinics in 2003. The rate of spread of the program has increased over time, with 25 total clinics using the MHI program in 2006 and 68 total clinics in 2008. Intermountain has also helped other organizations in Maine, Oregon, Mississippi, and Utah community health center clinics adaptively model the MHI program. Intermountain anticipates more than 120 clinics will be using the MHI program by 2009.

Intermountain has built in safeguards against growing too fast and losing control of the implementation processes. Intermountain uses learning organization techniques and works with existing institutional structures to support the implementation and spread of the program. There are ongoing meetings and opportunities for key players to meet, monitor progress, discuss encountered challenges, and learn from each others' experiences and practices, including monthly meetings and annual retreats. Partners in other states implementing the MHI model are sharing a standardized set of measures to provide meaningful outcomes comparisons and to advance the evidence base for MHI. Intermountain is very interested in understanding if other organizations can successfully run the MHI model and, if so, what they look like.

The MHI program is a comprehensive mental health approach that is available to all patients, not just those patients with disease-specific needs. Patients, and their families, complete a comprehensive assessment tool that investigates issues related to the full range of mental health concerns—depression, bipolar, anxiety, developmental concerns such as ADHD, and alcohol and substance abuse. This information is loaded into an algorithm that stratifies patients into mild, moderate, or severe categories and available resources are matched to the patient's level of need and preference. In general, physicians and nursing staff continue to provide care for about 80 percent of the patients in primary care based on established protocols and information feedback loops. The other 20 percent receive care from other specialized team members, depending on the need level and complexity of the patient's condition.

The comprehensive assessment toolset may appear lengthy and counter-intuitive; most mental health providers would say the families aren't going to complete the forms; but experience has shown that patients and families will complete the forms. The key is that the physician believes in the effectiveness of the toolset and how it provides insight into the patient/family situation. Physicians who are focused on the job will point out that the form will help patients understand and get the help they need. If it is coming from the physician, and the patient wants an answer to what has been a problem, they will fill it out. Adherence with the form has been remarkable. But the form was designed by clinicians with the guidance of behavioral health specialists and vetted by the National Alliance on Mental Illness (NAMI) as consumer friendly. In fact, the form has become almost a ritualistic tool that keeps the care team cohesive.

Care managers are generalists that carry the mental health perspective and skills across the medical disease spectrum as well. The mental health assessment and program is becoming the infrastructure for chronic care disease.

Team members use harm reduction strategies to improve education and to provide treatment for alcohol misuse; they facilitate involvement of families and community resources in social support and reinforcement of abstinence. Strategies that are tailored to the preferences of patients and communities are more likely to result in positive behavior change.

Unique to the Intermountain integrated care model is the inclusion of a family systems perspective. The patient and patient's family are listed first as members of the care team in Intermountain's patient literature. This idea is supported by a theory-based method and the training and tools, including a family pattern profile, for clinical team members to assess the family's style in dealing with stress and health problems and adapting the treatment approach to best mobilize the patient's family resources

Intermountain also includes outside resources as acknowledged team members in the patient education literature. Care managers make available to patients community resources such as NAMI and other community partners. NAMI has been an involved partner in Intermountain's MHI program development.

Within Intermountain, it is accepted that the implementation of evidence based medicine is the responsibility of the institution. The institution gathers the data and the evidence for best practices. The clinicians are responsible for implementing the best practices. It is the institutions responsibility to give the clinicians the resources and training they need in order to be able to deliver evidence-based medicine.

Using quality improvement techniques, Intermountain spent considerable effort developing measurement tools with graphic capability linking patient care processes with program and plan outcomes and costs in order to document outcomes and refine the allocation of services to the appropriate level of patient severity. The information is used to help build consensus among the various stakeholders and responsibly allocate resources to those patients for whom they can provide the most benefit.

One of the contributing factors to the success of MHI was the organizational housing of the mental health clinical integration system within the medical group. Since the MHI program is cost neutral, this placement made it possible for the nonfinancial justifications for the program to be recognized as important; physician satisfaction with the care they were providing, and patient satisfaction. Intermountain was not immune to the tensions between behavioral and medical health cultures, or to the concerns physicians initially felt regarding the new program—that they were “being made psychologists on the cheap.” With the attention to training, and time for the physicians to see how the program benefits their practice and the quality of care they deliver, physicians are now fully on board and asking for the program. Intermountain has also been partnering with the Institute for Healthcare Improvement to help with the processing of convening interested groups so that Intermountain can remain focused on care delivery.

This physician buy-in is essential to the program's success as well. The MHI program does require a redesign of clinic costs since the care manager is an addition to clinic staff. Behavioral health specialists may be clinic staff or may be financially supported by an umbrella department, depending on how the regional staff chooses to fund the program.

Program implementation is variable. Intermountain has identified core essential components, such as leadership, workflow integration, screening and clinical assessment tools, training, message logs, and registries with feedback reports, which are necessary to a successful program. Other elements are adaptable to the specific local environment of the clinic. Clinics are generally running in the black within 3 to 6 months, regardless of whether they function under staff or network models.

The cost neutrality of MHI for the health plan stems in large part from cost reductions in ER visits, psychiatric inpatient admissions and length of stay, and length of stay for inpatient admissions related to other medical conditions. These reductions are happening because people are getting the services they need with appropriately matched resources. Further, by effectively identifying and treating mental health issues, medical providers and care managers are taking the improved skills over into treatments for other chronic conditions. Since Intermountain is a fully integrated health care system, they can capture all the cost efficiencies.

Notwithstanding the above, billing, scheduling, and credentialing for the clinics is still a challenge for the clinics because the general financial reimbursement structure is still the perverse and fragmented structure all health care organizations face, and it often overwhelms the front staff. This is a factor that the CQI teams intend to address in the near future.

Intermountain is in the process of rolling the MHI program out to rural clinics. They have found their rural physicians have high mental health acuity; there is often no one else available locally to provide such care. While Intermountain has been exploring other work force solutions, such as mobile teams and telehealth care, rural physicians have already begun implementing some of the MHI tools. Even that limited contact with the program has demonstrated the benefits of the program to the rural physicians, and they are eager to hire additional staff and get the program up and running.

Integration for Intermountain is present when all systems are linked and standard processes are routinized and in place so that it doesn't matter who the team member is, the patients will get the treatment they need. MHI's sustainability is not an issue at this point. Integration has been institutionalized to the extent that Intermountain is past the danger point of killing the program by losing key leadership. The networks of involved clinicians and players have become self-supporting.

MaineHealth

MaineHealth is a nonprofit integrated health care delivery system serving 300,000 individuals in 10 counties in rural Maine which includes a provider network for the full care continuum, a public health component through a community health status program, community health education, and an integrated information system. MaineHealth also has a very robust quality improvement infrastructure, including the Clinical Integration Division, which is responsible for the development and piloting of clinical QI programs. MaineHealth had previously adopted the Chronic Care Model for all of its Clinical Integration activities related to chronic illness care, and has experience working collaboratively with practices, employers, health plans and patient advisory groups in improving care.

MaineHealth's history with integrating mental health and primary care began 6 years ago with the advent of their participation in the RESPECT-D trial funded by the MacArthur Foundation. Further grant support from the Robert Wood Johnson Foundation was used to disseminate the techniques and models further. As the experience with the program increased, leadership at MaineHealth committed to expanding the depression care program across the primary care practices associated with MaineHealth. The Institute for Healthcare Improvement “Learning Collaborative” approach was used for initial dissemination of the program. Practice outreach and electronic learning modules, supported by pay for performance programs, were used to increase the number of practices. Currently about 65 practices, roughly 80 to 90 percent, or 130 to 140 primary care physicians, within the system use tools for depression care developed by the program.

Concurrent with this process, MaineHealth leadership engaged in strategic planning and concluded that integrated mental health and primary care was a strategic priority. They recognized some shortcoming of the depression program existed that might be addressed by a broader mental health integration program:

• The program was only available to adults. The pediatric population did not have a similar level of quality of care for depression.
• The program focused on a disease-specific condition. Psychiatric comorbidities, such as anxiety and PTSD, were common but not addressed. As a result, some patients were not improving as expected because the comorbidities complicated treatment.
• While improvements were seen at the primary care practice level, improvements were also needed at the interface between primary care and the mental health system level to achieve the full potential of improvement of the provision of mental health care in primary care.

A presentation by Intermountain Healthcare was very influential to the decision process. The Intermountain MHI model appeared to meet all of the identified concerns. MaineHealth contracted with Intermountain Healthcare to help roll out the MHI model. MaineHealth also was able to modify a foundation grant proposal to support a pilot of at six primary care practices. MaineHealth is in the third year of that pilot.

The MHI model calls for care managers to go beyond a disease specific approach. MaineHealth already had care managers providing services for diabetes, heart disease, asthma, and depression, with about 25 care managers working in primary care practices. Many care managers had prior experience with depression; as a result of involvement in the MacArthur and Robert Wood Johnson Foundations funded activities with the organization. Having disease specific care managers take on an additional comprehensive mental health focus to their case load has been challenging. Staff at MaineHealth has provided support and education to the care managers to assist them in taking on this expanded role.

The MHI pilot has proven the importance of a clinician champion for a clinical improvement program. The champion PCP for one pilot practice has been on an extended sabbatical and program implementation has not been as smooth while he has been away. This location contracts with a local mental health agency for the onsite behavioral health service. The protocol in this practice calls for the PCP to hand a mental health assessment questionnaire to a patient with one or more suspected mental health diagnoses and the care manager follows up with collecting and scoring the instrument. Concerns about the risk of getting reimbursed for counseling services in primary care have served as barriers to the on-site behavioral health service. The use of the mental health assessment has been an uphill battle. PCPs at the clinic have not made use of the assessment and were concerned that patients were not likely to complete it.

The care manager at this location, a licensed social worker, had been involved in the roll-out of the depression care model. The PHQ-9, from the RESPECT-D program, is still used for annual visits and new patient screeners, as well as for tracking patients under care. It was her impression the patients have come to view the PHQ-9 as a regular process of care and a good fit for the standard patient. The MHI mental health assessment questionnaire is perhaps a better fit for the complex patient.

At another pilot location, one of the major champions is the licensed clinical social worker who functions as the onsite behavioral health specialist. At this location the behavioral health specialist is on staff and plays a more involved roll in the questionnaire followup and scoring and in patient treatment, including creating the treatment plan. In her experience, while some patients have needed more help than others with the comprehensive questionnaire, overall the patients have found that the ability the assessment questionnaire provides to self-evaluate is positive. It helps them put a label on the problem.

The behavioral health specialist uses the generalist approach to patient care required by the MHI and views it as professionally more satisfying than the more limited role she had played in the depression care program as a care manager. In turn, a PCP on staff finds the onsite presence of a behavioral health specialist positive as well. The ability to immediately hand off a patient in crisis and know that the patient will be helped is invaluable; it frees the PCP's time to be spent with other patients. The PCP also reported an increase in his general mental health knowledge base, the range of treatment options available, and comfort with identifying and treating patients. This PCP's opinion has been borne out by a provider survey which indicated strong satisfaction with the program.

Some clinics implementing the MHI program have been held back by the shortage of psychiatric practitioners, compounded by the rural location. One clinic currently hiring a psychiatric advanced practice nurse waited 2 years for the position to be filled after posting.

The payers are a tougher lot. MaineHealth has been accomplishing implementation with its own funds and help from grants, but the long-term feasibility requires bringing payers on board. Talks have begun with opinion leaders from other organizations to develop strategies to change licensing and reimbursement policies to remove barriers to MHI. Employers are also a potential focus for talks; they see the savings in disability, presenteeism and absenteeism when quality mental health care is provided, but it is still hard to get the payments to follow.

Until changes occur, sustainability is still a site by site phenomenon. Care management is funded by health system or clinic budgets, not payers. Revenues generated by providing direct mental health services may be adequate to support the cost of staff - at least, that has been the experience at Intermountain. Some pilot sites have the benefit of rural mental health licensure, which allows them to receive a higher reimbursement rate for mental health services. There is much more to learn about the financial sustainability of integrated mental health services.

Northern California Kaiser Permanente

Northern California Kaiser Permanente (Kaiser) began its integrated care in 1996 as part of a regionwide redesign of primary care based on Kirk Strosahl's model for integrated care. ⁶⁵ This redesign brought behavioral medicine, as well as health educators, physical therapists, pharmacists, medical assistants, and RNs, into a primary care team responsible for a panel of patients' total patient care. The team structure was designed to leverage valuable physician resources through physician extenders and to acknowledge the limits of the physician knowledge base. The system redesign comprehensively addressed deliverables, clinic structure, administration, and clinical processes.

The Behavioral Medical Specialists (BMSs), licensed clinical psychologists or licensed social workers, were co-located and functioned as generalist consultants for primary care visits, often used for unique primary care patient visits that involved primarily mental health concerns. BMSs adopted the culture of primary care and co-managed patients with regard to behavioral and emotional sequelae of primary care visits. The BMS helped with triaging of patients, difficult customers, somatizing, depression, and anxiety, and contributed to the population panel management with load management and scaling. Patients are generally referred to the BMS by medical providers with a warm hand-off, but patients can self- refer to a team BMS as well. Complex patients are seen in Psychiatry.

BMSs are supervised by their own subchiefs, but also have clinical and quality ties with the Department of Psychiatry. Administrative supervision is handled through the clinic.

Patient information is shared through an EHR which is generally available to all providers. Patients under the care of a BMS are notified and agree to the fact that information will be shared with the physician and includes charting of behavioral symptoms and issues. The behavioral charts are not open to medical assistants. If a patient is referred to Psychiatry, the patient will be asked permission to share information with the primary care physician, and prescription information is always available. However, Psychiatry has confidential notes not available to other providers.

Co-location has been critical to the success of the program. The convenience and lack of stigma has helped overcome the “referral to no services” when patients wouldn't cross the bridge to mental health specialty services because of stigma and lack of convenience. Patients also feel more comfortable coming into their familiar primary care environment and being treated by a team member that has already been identified. This has resulted in improved access to mental health care for the patient panel.

Likewise, for providers, co-location has resulted in cross-fertilization between providers of different disciplines. Consultation with BMS staff has improved the quality of care, and chronic care management specifically. Greatest results were seen in patients who aren't progressing or aren't adherent to a treatment plan.

While several components of the initial redesign were eventually dropped, the behavioral health component has continued and BMS staff are currently playing an integral role in the implementation of a clinical improvement program for systematic depression care management based on the IMPACT model.

Kaiser participated in the IMPACT study and found that even with co-located BMS staff, members treated according to the IMPACT protocol showed significant improvement in depression outcomes when compared to members treated by BMS staff that did not implement IMPACT. Kaiser learned that systematic monitoring and followup provided an additional impact on patient outcomes. The BMS model is a generalist approach, created to address a variety of common mental health conditions seen in primary care, with short targeted interventions. What was missing were the tools for specific tracking and monitoring of patient progress towards improvement and remission. Prior to IMPACT, each primary care team would have their own decision process regarding diagnosing, measuring, and tracking patient care. The new clinical improvement project for depression care allows systemization across sites.

A second major force driving the clinical system improvement project was the money being left on the table through coding inefficiencies. During the work with IMPACT, Kaiser discovered variability in the way that clinics coded mental health and behavioral health services. PCPs were not diagnosing specifically enough, nor were behavioral health services coded specifically enough, and Kaiser was missing opportunities to maximize revenue from Medicare reimbursement for Major Depression as a risk adjusted condition. The ability to increase revenue was certainly a selling point to management for investing in system redesign.

The new clinical improvement project also involves the introduction of new tools to a data warehouse, which will be used to help refine the targeting of the systematic care process. The PHQ-9, and other outcome reporting, allows integrating depression care data fields into a population management IT system. The focus on data will include patient contacts, how contacts are coded, use of the PHQ-9, and eventually outcomes data. In time Kaiser expects to be able to share outcome data with employers to demonstrate effectiveness of depression treatment, as well as reducing absenteeism and presenteeism. New electronic depression treatment tools for patients are also being developed. This would be in keeping with Kaiser's history of providing rich resources to patients for education and self-management skill development (which are generally at no additional cost to members).

Staff at Kaiser feel the history of the BMS program translates into an advantage which will allow more rapid program implementation. Site specific expertise and institutional memory around the collaboration of behavioral and medical providers for mental health conditions are already in place.

Even with this history, though, change is still difficult for providers. There is a learning curve for BMS staff in adapting to the medical model and systematized processes of care are particularly difficult for those who were most used to the freedom of treatment options found in more traditional mental health models. Some have chosen to leave the position because it wasn't a comfortable way of working. The traditional psychotherapy model is dynamic, and the therapist is ultimately responsible for the patient's care. In a consultant model, a major role of the BMS is as educator. Systematized care, for some, may feel even more confining.

The clinical system improvement is being implemented region wide. There are over 80 depression champions identified across the system helping with training and providing expertise.

Kaiser is making some changes to the IMPACT model. With generalist BMS in place, the functions of the depression care manager are being distributed across team members. Nor will they be adopting the specific PST used in the protocol, allowing the BMSs to continue functioning as they are trained to do. No new staff will be required, with the possible exception of population management assistants (medical assistants) that assist with patient panel management. It is anticipated that the increased patient workload will be offset by the more efficient systematic processes. The population management assistant leverages the BMS (who was originally brought on to leverage the PCP).

The new clinical system improvement program will be rolled out on new index cases of depression beginning antidepressant medication, and then will be expanded to all adult patients with depression. Beginning with a defined population will allow Kaiser to test and refine the system and allow the providers to develop familiarity with the systematic care process. Later expansions will include other high risk populations such as OB/GYN for post-partum depression or domestic violence, or patients with diabetes. Screening will only be added at some later time when the organization is confident in the program and organizational capacity.

Minnesota DIAMOND Initiative

The Minnesota DIAMOND Initiative is an evidence-based care management program that provides systematic and coordinated care for adult patients with major depression in primary care settings. The program was built on Wagner's Chronic Care model and IMPACT study protocols. Key care elements include assessment and monitoring with the PHQ-9, use of a registry for systematic tracking, formal stepped care protocols and relapse prevention. Nurses, medical assistants, or people with a clinical mental health background in a depression care manager role, perform the care functions, meeting weekly with a consultant psychiatrist for designated case review meetings. Specific duties of the care manager include patient education, self-management support, coordination of care with primary care and behavioral health providers, and facilitating treatment changes identified by stepped care protocols. The care managers also facilitate communication between the mental health and primary care providers. Some care managers receive additional training to provide PST, a brief solution-focused treatment with efficacy for use in the primary care setting.

The DIAMOND program is being rolled-out in several waves over the next 2 years. The first wave was implemented in five medical groups with ten clinics. The staggered waves were constructed to allow time for adequate clinic staff training and preparation as well as ramping up for the payment redesign model. The Institute for Clinical Systems Improvement (ICSI) also functions as a certifier of clinics for readiness to implement and certifies care managers upon completion of their training.

DIAMOND's development involved a collaboration of medical groups, health plans and payers, governmental bodies, and consumers, overseen by a steering committee comprised of major stakeholder representatives and facilitated by ICSI. ICSI is an independent organization that facilitates development and implementation of evidence-based practices for its 57 member medical groups and helps organizations build quality improvement structures, systems, and culture. ICSI represents about 85 percent of Minnesota's physicians, with funding support from six major Minnesota health plans. While the health plans fund the organization, governance is conducted through a board comprised of 11 members from medical groups, and three from the health plans. This, along with independence from governmental or political bodies, allowed ICSI to be perceived as a trusted independent body whose actual constituency is the patient and quality patient care.

A major key to this initiative was the redesign of payment structures to accompany and support the redesign of care processes. All plans abide by the same payment and service protocol. A care management fee was instituted and is payable to medical groups that are participating in the DIAMOND project for certified care managers following DIAMOND care elements and protocols. The care management fee covers a specified bundle of services billed for using a single service code and is paid monthly. Provider/patient visits, both medical and mental health, are billed separately. Care managers and psychiatry time is a fixed cost to the clinic, so there is incentive to keep the caseload full.

The payment structure redesign allowed the DIAMOND project to avoid major barriers encountered in more limited initiatives that found physicians would not commit to depression care management programs at an active level if the program was not available to the majority of patients. By bringing on board the majority of payers in the metro area, physicians did not have to be concerned about differential treatment for patients, multiple parallel care processes, and financial support for activities that have been otherwise unbillable.

ICSI invested considerable effort in providing the medical groups with detailed planning and implementation materials so that reliable cost information was available for the medical groups. Similarly, ICSI worked out a recommended standard process for the actual employees responsible for coding and payment at the health plans.

There was a considerable learning curve regarding anti-trust concerns during the process, and it is a major concern. All contracts between each medical group and payer were negotiated individually, thereby forestalling anti-trust concerns regarding price setting, but common elements were included.

ICSI addressed the leadership barrier by requiring strong commitments of local champions for each participating medical clinic. Participating organizations had to ensure to ICSI that key decisionmakers would be directly involved in the planning process. Authority and accountability had to accompany commitment. Also, each group had to promise and deliver a lead physician who would champion the initiative in the clinic. In addition, ICSI required the participating health plans and payers to sign letters of commitment to the payment redesign. Signators were required to hold positions of responsibility and authority necessary to provide follow through.

The health plans fund ICSI, but they are not responsible for the major governance. Board membership draws from diverse stakeholders. ICSI is also not a political body or affiliated with a political party. The nature of the organization, neither politically affiliated nor perceived as being a “puppet” for the health plans, is why ICSI is trusted as a facilitator of collaboration. It is the role ICSI plays for the community of varied stakeholders that ICSI values most. If any representation exists, ICSI represents patients and patient care.

Roll-out of the program was staged in phases to give ICSI, the health plans, and the DIAMOND initiative, a controlled process in order to apply learning organization skills, adjust and adapt materials from lessons learned, and adequately support the process. The cautionary side to the involvement of buyer groups in the initiative, including the strong support they have provided to the process, is that the excitement tends to drive a push to expand and speed up the roll-out process and make it bonusable. If the initiative moves too quickly, there is the danger that the program gets diluted and won't be able to demonstrate effectiveness.

ICSI has also been working collaboratively with organizations to assure that measurement of process and outcomes is in alignment with evidence based quality depression care. ICSI has been working with Minnesota Community Measurement, a nonprofit organization working to improve health by publicly reporting health care information, and, along with other organizations, the National Committee for Quality Assurance (NCQA), to improve quality measure for depression care. ICSI has also been working with organization such as the Buyers Health Care Action Group (BHCAG) and their Bridges to Excellence (BTE) program in order to provide a pay for performance program for high quality depression care in the state.

Veterans Administration

The VA mental health initiative focuses on serious mental illness and depression. The VA's approach to integrating primary care and mental health has benefitted from a number of initiatives that have been sequentially and cross-sectionally coordinated. One is tempted to describe this approach as acronymistic; each project has its own acronym. However, the scope and trajectory of each project has been coordinated to ensure that momentum is maintained and each component builds on its predecessor.

The VA has utilized modern CQI techniques combined with principals of evidence-based medicine to introduce and maintain this concept. The projects are individually tailored to each site but have some core components that include leadership support at the Veterans Integrated Service Network (VISN) and medical center level, creating a context of collaboration with local leadership, problem identification and intervention planning, and team building; evidence-based guidelines and tool kits; education; and tailored informatics that includes tracking software and patient registries.

Projects are unique and labor intensive. The basic model is tailored to the individual needs and constraints of each site but consists of a primary care team, a depression case manager who provides active engagement, proactive followup, and immediate specialist (psychiatric) consultation when a problem arises

The VA has been working on integrating primary care and mental health around care for major depression for some time. The effort built on the Katon collaborative care model, and its initial quality improvement version under the Partners in Care project. Partners in Care assisted six medical care organizations (MCOs) (48 primary care practices) in improving depression care. The MCOs were willing to support integrating primary care and behavioral health OR creating better payments methods for cognitive behavioral therapy, but not both. The study thus included two intervention arms: (1) minimal care management followed by encouragement to access CBT with decreased copay and (2) 6 to 12 months of care management in primary care. In both arms, researchers trained expert leaders and care managers from the MCOs to implement the study intervention. These leaders in turn trained clinicians in the practices. Researchers had no direct hand in implementing the intervention in the practices. This effort produced more positive outcomes than earlier CQI models, but the positive results but did not endure.

The Mental Health Awareness Project followed Partners in Care as a CQI project at Kaiser and the VA. It used resources developed by Partners in Care and Katon to provide resources and consultation to local teams who developed their own QI agendas. It operated under two models: (1) a local team helped by central experts and (2) regional leaders who played a more active role in improving care. Reviewers tended to see the project as a negative trial because it did not improve depression symptom scores. It did increase patient satisfaction with care, however, and where the intervention care model developed by the teams was at least minimally evidence-based, depression symptom scores significantly improved across all depressed patients cared for by the practice. It may be worth noting that in this study, as in two prior CQI studies, but unlike the Katon, Partners in Care, and other collaborative care intervention studies, the representative patients participating in the evaluation received no individual interventions beyond participation in a survey; their clinicians and practices were blinded as to their participation in the study. These patients only experienced improved depression care under whatever circumstances patients similar to them who attended study practices experienced it.

Meanwhile, meta-analysis showed that collaborative care interventions were effective and cost-effective, based on over 35 randomized trials and over ten cost effectiveness analyses. The VA team next sought to use the evidence base from these trials to help VA regions create something that was intrinsic to the VA. They got a 2-year grant to create and adapt tools to the VA setting. They worked with VISNs, using expert panels of VA regional leaders to decide how to implement the evidence base on collaborative care for depression in VA. They then assisted these regional leaders in organizing and implementing the intervention features decided upon in places identified by the panels. This project became TIDES (Translating Initiatives in Depression into Effectives Solutions), which involves using case managers and treatment protocols to assist primary care clinicians in managing depression, often offsite by telephone. Care managers are backed by mental health specialist review and consultation, enabling patients who require or prefer specialty mental health services to access them.

TIDES was continued as RETIDES (Regional Expansion of TIDES). It was organized as a bottom up national implementation. The RETIDES evaluation was based on performance measures derived from the EHR and a provider survey, and is ongoing. It ends in the fall of 2008. In 2006 the TIDES intervention was picked up under funding from the Office of Primary Care and Mental Health Integration, facilitating further spread. Care managers have rotating panels of 75–100+ patients at any time. Fifty care managers have been trained by TIDES; 38 are currently working. At least 17 medical centers (containing 50+ primary care practices) in seven VISNs have an active TIDES program. Most TIDES centers got mental health/primary care initiative grants to maintain the care manger funding. One TIDES VISN and five practices discontinued TIDES because of staffing and/or funding issues.

In implementing a program on this scale, especially one that relies on local initiative, it is easy to lose control. Sites and the program are influenced by what is happening locally and within the VA. The mental health/primary care initiatives tried to do a lot fast, and were not set up to provide training to new sites. There is a lack of performance measures geared to tracking critical TIDES components. As a result, there is likely to be substantial resulting variation among post-RETIDES sites in exactly how collaborative care is implemented.

The Quality Enhancement Research Initiative (QUERI) has funded the bridge component of RETIDES, which is designed to learn how to implement TIDES and implementation programs like it across the VA. The bridge project has found that funding is NOT the most critical barrier to implementation in the case of mental health/primary care integration. There is increased funding for this as a result of concern about the impacts of the war in Iraq and Afghanistan and resultant PTSD.

Funding from the Office of Primary Care and Mental Health Integration will support the establishment and added operational costs of these programs. The duration of this finding is unclear since it was congressionally mandated, but there is a belief that once the programs become established they will be maintained. Of the $35 billion budget for the VA medical care, $2 billion (now $4 billion) goes to mental health, with a special set aside for mental health in primary care. This should become a recurring funding program. VISNs are currently flooded with more money than they can spend on mental health.

Despite all the effort and attention, implementation is described by one commentator as like slogging through molasses. Despite efforts at integration, mental health and primary care culture may clash at individual sites. Barriers to implementation (and ultimate incorporation) include clinical inertia (clinician reluctance to modify practice style or a course of treatment); a lack of recognition about depression, which has been offset by mandated screening measures and publicity from currents veterans' mental health (especially PTSD); and time constraints. Performance measures could re-enforce what is being introduced, but they do not. They drive the practice; doctors work to achieve mandated tasks; unfortunately, the mandatory screening is not always appropriate. Ironically, there may be too many case managers at times. A patient with complex illness and multiple comorbid conditions may have a case manager for each diagnosis and for eligibility issues as well. As one observer facetiously put it, they may need a case manager to coordinate all the case managers.

One aspect of mental health/primary care integration that has been virtually impossible to achieve to date is primary care-based, evidence-based (manualized) CBT. Doctors are often quick to use drugs because they are fast and easy. Partners in Care showed the enduring effects on patients (now shown to persist over 10 years) of enhanced access to primary care-based CBT for patients who prefer or need it. There is a need for both more primary care-based and more mental health specialty-based psychotherapy and CBT.

Although the VA is spared some of the financial issues that haunt a fee-for-service payment scheme, it has other pressures. Care is judged by productivity criteria that may not capture important elements of depression care and may create disincentives to nonpharmacologic approaches. Quality measures include access time to get appointments and waiting time. Quality measures require that many tasks be performed.

Performance measures can be a problem. Using quality performance measures from the civilian world may not fit VA style e.g., followup of depression by office visit rather than telephonically, although the latter is just as effective. As a result, doctors feel harried and busy. An additional reason why civilian measures may not be readily applied in the VA is that patient complexity is higher than in most civilian settings.

Ironically, the client inertia blamed for the difficulties of getting the program established may help to sustain it. Sustainability will depend on sustainable habits (positive side of clinical inertia). Habits based on integration are now in place. Moreover, the shortage of psychiatrists will prompt this model, because it uses care managers aggressively.

Aetna — Depression in Primary Care Program

Aetna has invited all its primary care physicians to participate in the depression in primary care program, designed to improve the care of depression. All doctors need do to be eligible is to make available some time to talk about the program. In addition there are voluntary training materials available online (www.aetnadepressionmanagement.com). They are tailored separately to physicians and their office managers. CME credits are available, with two free CEUs offered to physicians.

The basic collaborative care model, which began implementation in 2005, has three components (with the patient at the center). It is based on materials developed at Duke and Dartmouth, as well as IMPACT and RESPECT-D.

1)

Physicians screen for depression using the PHQ-9. They are instructed to ask the first two questions and continue only if they get positive responses to these two questions.

2)

Aetna care planners/case managers phone patients identified and referred by the practices at 1, 4, and 8 weeks after treatment to ask about their understanding of their treatment and any problems they are encountering. They administer the PHQ-9 at 4 and 8 weeks. More frequent calls are made as needed. Copies of the PHQ-9 are sent to the physician prior to the next visit with the patient. Special alert notes are made if the patients are not improving or getting worse.

3)

Behavioral health referrals are facilitated but made only at the physicians' behest. Care managers assist with these referrals when requested. Physicians can also consult with an Aetna psychiatrist whenever they wish.

Aetna works exclusively with contracted participating physician practices. Of their 200,000 contracts, only about 20 percent of these practices would be suitable for the program. Dr. Un estimates that about 5,000 practices have the organizational infrastructure necessary to support successful implementation. The essential organizational components include:

1)

An organized quality improvement process.

2)

The capacity to track data.

3)

An electronic medical record is not a requirement to participate in the program, but it could be a plus, if it is flexible enough to interface with their system.

4)

The office management infrastructure is the key component: they must be able to handle the special workload imposed by identifying and dealing with this subset of Aetna enrollees.

Physicians receive a welcome kit that includes copies of the PHQ-9 and instructions on how to administer it, referral forms for mental health consultations, members information about the program and benefit, and information on how to submit claims for the screening. Cooperation of office managers and staff is seen as key to the success of the endeavor.

Physicians are reimbursed for completing the full PHQ-9. They bill as contracted for ongoing depression care; there are no special fees or other added payments. The physicians' additional costs are largely tied to screening.

Aetna has offered to send physicians a list of patients with comorbidities that suggest they may be at higher risk for depression screening targets but few physicians have taken up the offer.

It is much easier to get practices to conduct the screening than to conduct the screening than to get them to refer cases to the care managers. A major barrier to successful implementation is centered around the problems of integrating the process into the practice workflow. Indeed, this process seems to work best if a practice uses the screening for all patients, not just Aetna enrollees. The major barriers to successful implementation include:

1)

The need to identify Aetna members who make up 20 percent at most of all patients in a practice (usually much less).

2)

Need to administer the PHQ-9 (a break in routine).

3)

Need to submit a special claim form using a billing code developed specifically by Aetna for the screening.

Of the approximately 5,000 practices that were approached to sign on, about half agreed but there was considerable drop off, especially in follow on after screening.

It is hard to implement this program with a single payer when practices work with many carriers. It requires too great a special routine. Large practices seem to have the administrative staff to cope with the special processing better than small ones, which are inundated with programs from many carriers.

So far, the evidence of impact has been seen in improved PHQ-9 scores. A study of the medpsych case management program reported at Academy Health in 2006 suggested that that program did save considerable money for a targeted group. They showed a decrease in medical costs of $175–$222 PMPM (most of this in inpatient care) and an increase in pharmacy costs of $21–$40 PMPM (only $8–$11 in antidepressants). The net savings was about $136–$201 PMPM. However, these figures were limited to a small subset of Aetna enrollees who had very high risk of medical care and were already in an active case management program; they also had higher risks of depression. These results led to the decision of implementing the depression in primary care program.

While the return on investment (ROI) for the case management program was estimated at 3:1, work is underway to estimate where the ROI for the depression in primary care program. Aetna has created its own risk predictor system (PULSE); they estimate that a score of nine or more is the tilting point.

A major barrier in implementing this program more widely is its idiosyncratic nature. Practices must set up a separate work flow for Aetna clients. Aetna would like to see the approach adopted by more plans to improve the work flow and increase the likelihood of operational implementation. They are collaborating with a pilot program in New York City to promote wider adoption by health plans and sponsor work by the Carter Center to encourage integrated care for depression.

There is some sense of a culture change to become comfortable with this new approach to care; about a third of practices are comfortable, another third are not, and the remainder are open to talking about it. Apparently no specific time frame or criteria have been set to determine the success of the enterprise, but they will continue to look at it as a program that takes time to become incorporated.

Aetna operates its own pharmacy benefits manager (PBM) program and uses that information in the depression program. About 90 percent of HMO members and 60 percent of preferred provider organization (PPO) member have the Aetna PBM system. They use data from that system to identify high risk patients for practices to screen. They have a comprehensive algorithm that includes the use of antidepressants. Basically they want practices to have a high rate of positive screens to encourage them to screen and act. They use the PBM information to flag patients who fail to fill their first antidepressant prescription and those who do not refill at 3 months. This information goes to the case managers and the primary care physicians.

Aetna plans to extend this approach to integrating behavioral health into primary care to include a program for screening, brief intervention and referral for alcohol abuse. Their prime target is problem drinkers rather than active alcoholics. They will use AUDIT (an alcoholism screening tool developed by the World Health Organization). They will encourage practices to use brief interventions including medical treatments, but recognize that many patients will require care from alcohol counselors as well.

Corphealth

Corphealth, soon to be branded LifeSynch, manages the Integrated Medical and Behavioral Health (IMBH) program for Humana. The program has been in operation for a little over 1 year. The main focus is case management. The case managers provide telephonic coaching and support (including facilitating conversations to deal with their emotional issues and to assure they are receiving the right kind of care) to Humana subscribers who have been identified by various screening methods. Primary care physicians are notified of what is occurring but are not actively involved. The integration occurs at the case manager level. Behavioral health case managers interface with medical case managers through electronic and telephonic means. When enrollees enter Humana they are sent a health risk assessment (HRA) form to complete. This HRA contains screening questions that identify persons at potential risk of behavioral problems (broadly defined). This screening is augmented by a claims process review that flags high risk comorbid conditions. Persons with an inpatient admission and those with a diagnosis of chronic pain are also screened in. Persons who are screened for and consent to IMBH case management services are contacted by a case manager who talks with them to assess the extent of any behavioral problems. The case manager develops a comprehensive care plan, and may take on a coaching role modeled after Prochaska's Stages of Change model to address obstacles to making indicated behavioral changes, facilitating a change in behavior, and monitoring the outcome. Much of this is symptom management. If the patient appears to suffer from a significant behavioral problem, the case manager will refer to a mental health specialist. Most primary care practices are judged not to be able to, or be interested in, managing behavior problems; no cognitive behavioral therapy capability tends to be available. If medications are prescribed the case manager will work on adherence, per the enrollee's care plan.

Case management may also be triggered by prescriptions for medications to treat serious mental illness identified through the Humana PBM system. The PBM system may also alert case managers when a mental health medication is not refilled on time to alert them to focus on adherence. In addition to PBM alerts, case managers will also ask patients if they suffer from any serious mental illnesses.

The case managers include both the behavioral case managers employed by Corphealth and the medical case managers employed by Humana. The former are primarily mental health clinicians; the latter are nurses. The Corphealth case managers are trained in techniques to destigmatize mental health and behavioral problems. They employ sales approaches to make their contacts less clinical and off-putting for patients. The initial intensive training, covering both systems and methodologies, lasts a month. There is ongoing training through case conferences and other feedback. Humana nurses receive opportunities to attend training regarding behavioral health topics and resources, as well.

They have received almost no complaints from providers. This program requires almost no active participation from physicians. They are notified when a patient is referred to a behavioral health specialist but need not do anything active. Pains are taken not to make actions seem accusatory. This program provides assistance to providers, so there is little negative reaction.

Thousands of Humana beneficiaries have been screened. Less than 50 percent of those getting case management are referred to behavioral health specialists. At present Corphealth provides case management for 3,125 enrolled members.

Corphealth monitors the effectiveness of this program by tracking changes in medical spending, especially hospitalizations and emergency room visits. They are interested in the ROI. Although the program is still in the midst of robust development, leadership believes they have seen a positive ROI, and are making efforts to produce outcomes that effectively showcase the product and process. Plans are underway to also track patient satisfaction through surveys sent after completion of individualized program goals.