Capstone Project- JOHNS HOKPINS MEDICINE

D e v e l o p i n g a n I n t e g r a t e d P r i m a r y C a r e P r a c t i c e : S t r a t e g i e s , T e c h n i q u e s , a n d a C a s e I l l u s t r a t i o n

m

Barbara B. Walker Indiana University

m

Charlotte A. Collins Geisinger Medical Center

Numerous studies have now demonstrated that integrating behavioral

health and medical care can reduce medical costs, improve patient

and provider satisfaction, and enhance clinical outcomes. Given this,

one might expect that behavioral health programs would be fully

integrated into primary care clinics across the country, but in fact

integrated primary care programs remain quite rare. One reason for

this discrepancy is that implementing such programs has proven to be

extraordinarily challenging. Most of the integrated programs that are

currently operating successfully are in settings where professionals

are all members of the same health care system (e.g., HMOs, the

Veterans Administration, Departments of Family Practice, etc.). Many

providers, however, are in communities where various services are

provided in different locations from different organizations that have

very different clinical, administrative, and financial structures. In these

situations, the challenges are even greater. The authors describe a set

of strategies and techniques providers can use to move their health

care system toward a higher level of integration and illustrate how

they applied these steps to develop and assess the impact of an

integrated primary care program in the state of Rhode Island. & 2009

Wiley Periodicals, Inc. J Clin Psychol 65:268–280, 2009.

Keywords: integrated care; collaborative care; primary health care;

integrated services; delivery of health care; mental health services

Correspondence concerning this article should be addressed to: Barbara B. Walker, Indiana University, Department of Psychological and Brain Sciences, 1101 E. 10th Street, Bloomington, IN 47405; e-mail: [email protected]

JOURNAL OF CLINICAL PSYCHOLOGY, Vo l . 6 5 ( 3 ) , 2 6 8 – 2 8 0 ( 2 0 0 9 ) & 2009 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com). D O I : 1 0 . 1 0 0 2 / j c l p . 2 0 5 5 2

Introduction

In the early 1960s, physicians at Kaiser Permanente Health Plan noticed that the majority of primary care visits were from patients who were found to have no organic pathology. Results of their 20-year longitudinal study (Cummings & VandenBos, 1981) revealed that 60% of visits were from patients who had no physical disease, and the vast majority of these patients suffered from depression, anxiety, stress, and unhealthy lifestyles that negatively impacted their physical health. In response, this Health Maintenance Organization (HMO) began to apply the biopsychosocial approach (Engel, 1977, 1980) by integrating behavioral health

1

and medical services in their primary care clinics. A series of subsequent studies found that savings in medical utilization exceeded the costs of providing behavioral health treatment (Cummings & Follette, 1968; Follette & Cummings, 1967). More recently, integrated primary care programs have been developed in other

systems as well. The Veterans Administration (Druss, Rohrbaugh, Levinson, & Rosenheck, 2001; Elhai, Richardson, & Pedlar, 2007; Hedrick et al., 2003), the Air Force (Runyan, Fonseca, Meyer, Oordt & Talcott, 2003) and other branches of the military have all been investing heavily in developing integrated care programs. Taking all the evidence together, Blount (2003) recently concluded that in certain populations with certain types of patients and problems, integrating behavioral health and medical care can reduce medical costs, improve patient and provider satisfaction, and enhance clinical outcomes. Given these findings, one might expect that behavioral health programs would be

fully integrated into primary care clinics across the country. Unfortunately, this is not the case; integrated care programs remain extremely rare. One reason for this situation is that the health care system in the United States is not designed to foster the development, implementation, and/or maintenance of integrated services. Our largely fee-for-service structure provides strong incentives for performing medical procedures and doing diagnostic tests but few, if any, for focusing on prevention, communicating with other providers, and/or coordinating care. In addition, behavioral health and mental health have not yet achieved parity with ‘‘physical’’ health despite significant efforts to bring about this change. As a result, behavioral and mental health factors are often ignored clinically and ‘‘carved out’’ financially, resulting in fragmented, poor-quality, more-expensive care. This, combined with the misaligned financial incentives, often leads to insurmountable barriers for those attempting to develop any type of integrated care program.

Given this structure, it is not surprising that most integrated programs to date

have developed in settings where professionals all take care of a given population

within the same health care system. Examples include HMOs, Departments of

Family Medicine, the Air Force and other branches of the military, and the Veterans

Administration. Many communities across the United States, however, have few (if

any) such unified systems, making integration even more challenging. Given the

challenges that occur even when all providers work within one system (Blount, 1998),

how can care be integrated in communities where primary care teams, psychiatry,

psychology, behavioral medicine, and training programs all operate within different

systems in different locations with completely separate administrative and financial

structures?

1 In this article, we use the broad term behavioral health to include both mental health and behavioral

medicine services.

269Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

In this article, we present a more detailed rationale for integrated primary care, and then describe four specific steps that can be taken to move one’s health care system toward a higher level of integration. We then illustrate how we applied these four steps to develop an integrated primary care program that linked several separate systems together in the state of Rhode Island: a Department of Behavioral Medicine, a clinical psychology training program, and a private fee-for-service primary care office. Finally, we present some data assessing the impact of the program.

Why Integrate Care?

According to Selden (1997), most health care plans spend only about 4–6% of their annual budget on behavioral health care, suggesting that it may not be worth expending much effort in this area because it is inconsequential. This figure is misleading, however, because most behavioral health care is actually carried out in the medical sector. The majority of people seeking help for psychological problems are seen by their primary care physician (PCP) and not by a mental health specialist (Regier et al., 1993). Not surprisingly, the primary care sector is now often referred to as the ‘‘de facto mental health system’’ (Regier et al., 1993). Patients with chronic medical disorders are more likely to suffer from

psychological disorders than those without any medical conditions (Wells, Golding, & Burnam, 1988). The most prevalent disorders among this population were found to be depression, anxiety and panic, somatization disorder, and alcohol abuse. Compared to the general community, patients with medical disorders are two to three times more likely to suffer from major depression (Regier et al., 1993). Panic disorder and somatization disorders are 10–20 times more frequent in primary care settings (Katon & Roy-Burne, 1989), and substance abuse disorders are three to five times more common in a primary care practice than in the general community (Regier et al., 1993). Patients with comorbid physical and psychological disorders are extremely costly

for the health care system. Studies at the University of Washington in Seattle found that a relatively small percentage of patients (10%) accounted for 29% of all primary care visits, 52% of specialty visits, 40% of in-patient stays, and 26% of all prescriptions. Among these high utilizers, 50% were psychologically distressed (Katon et al., 1990). Other studies have demonstrated significant costs associated with each of these types of patients. For example, patients with somatization disorder were found to use nine times more overall health care services than other patients (Smith, 1994), and depressed patients were found to use three times more services (Katon & Schulberg, 1992). Also, patients with anxiety and panic disorder have 10 times more visits to the emergency room than those without anxiety, and 70% of these patients actually see 10 or more physicians before they are accurately diagnosed with anxiety (Katon & Roy-Burne, 1989). Health care costs for families with an alcoholic member are twice that of families without alcoholism (Holder & Blose, 1986). Why be concerned with behavioral healthcare in a primary care setting? One

reason is that patients typically present to primary care providers with all their problems, and these problems are not purely medical; they usually include biological, psychological, and social components. In a classic study, Kroenke and Mangelsdorff (1989) found that less than 30% of symptoms seen in primary care were classified as having an identifiable organic cause after one year. There remains little doubt that psychosocial and lifestyle factors play a significant role in chronic illnesses and

270 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp

somatic complaints. Ignoring these factors is costly to the system and leads to inefficient and ineffective patient care. Although PCPs are inundated with patients who require behavioral health care, behavioral health providers have developed cost- effective treatments that reach only a small fraction of those who could benefit. The challenge is to find ways to provide quality, integrated care to improve the current system. Changing any system is challenging, but changing the way health care is delivered

is particularly challenging given our current health care system. Not only are incentives misaligned, they are misaligned in different ways in each state, region, and community. As a result, integrated programs developed in one community may not necessarily succeed in another. There are no simple formulae for developing or maintaining integrated care programs. Local and regional differences significantly impact the configuration of service delivery, and understanding these differences and targeting an intervention to a particular area and specific setting is critical. Fortunately, there is a positive side to this challenging situation. Valuable lessons

have been learned from both successes and failures in many different types of systems in different regions across the country, and many of these have now been published (see Kessler & Stafford, 2008 for an excellent example). Through these efforts and those of professional organizations dedicated to fostering integrated care (e.g., Collaborative Family Healthcare Association and Society of Teachers of Family Medicine), it has become clear that the process of developing integrated programs is critical, and it is now possible to identify some general strategies that have been successful. Below we propose four specific steps to guide the process of developing an integrated care program. In many ways, these steps parallel the clinical process used to help individual patients change: (a) analyze the situation and forge trusting relationships; (b) collaboratively set realistic goals; (c) identify and find ways to overcome the barriers to change; and (d) implement a plan, test the outcome, and revise the plan.

Four Steps Toward Integrated Care

Step 1: Analyze your health care system, identify potential collaborators, forge, and strengthen alliances.

A helpful first step is to begin analyzing your health care system by generating a list of all potential collaborators in the community. When compiling the list, it is important to think broadly and consider all those who refer patients to you, all those to whom you refer patients, your colleagues, and the many different types of institutions in your community (e.g., hospitals, HMOs, university training programs, medical schools, etc.). After developing the list, it may help to draw a detailed diagram delineating exactly how all the various individuals and groups currently interact. Give careful thought to which people and/or groups might be motivated to develop a closer partnership. The importance of personal relationships in this process cannot be over-

emphasized. Every conversation with another health care provider provides an opportunity to begin forging alliances in an effort to move toward a higher level of integration. It is particularly important to collect information about what difficulties potential collaborators encounter on a daily basis. For instance, asking, ‘‘What clinical problems frustrate you most in your practice?’’ can lead to a fruitful discussion of how an integrated care model might improve the situation. Many

271Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

PCPs, for example, describe having great difficulty managing patients with problems such as somatoform disorders, mood disorders, chronic pain, substance abuse, and obesity. An integrated care program could better manage these issues and benefit everyone involved. Despite the fact that clinicians typically prefer to focus only on clinical issues, the

clinical, financial, and administrative spheres all need to be considered when developing integrated care networks. Failure in any one of these spheres will lead to overall failure of an integrated program (Patterson, Peek, Heinrich, Bischoff, & Scherger, 2002; Peek & Heinrich, 1995). As such, in addition to working on relationships with other clinicians, it is equally important to develop good working relationships with nurses, technicians, billing personnel, office managers, and anyone else involved with patient issues. Without support from everyone involved, change is very difficult to achieve. Within the potential collaborators identified, consider which individuals might serve as ‘‘champions’’ for the cause. Individuals who see the merit of enhanced collaboration are likely to have greater success convincing others within their own ‘‘system’’ of its worth than would an individual who is outside their system.

Step 2: Assess where you are now on the continuum of integrated care and set realistic goals for change.

It is important to assess exactly where you fall on the continuum of integrated care before making any changes. As Blount (2003) has pointed out, a continuum of care exists with regard to both location and the nature of the relationship between providers. At one end of the continuum are situations where behavioral health providers work in different locations than the primary care providers, have completely separate treatment plans, and do not communicate with one another. At the other end of the continuum are providers who communicate regularly and share the same office, the same administrative staff, the same billing system, the same charts, and the same treatment plans. A worksheet illustrating this continuum is

Figure 1. A worksheet illustrating the continuum of care that exists with regard to both location and the nature of the relationship between providers.

272 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp

shown in Figure 1. The worksheet is designed to help those developing programs identify where they are on the continuum. After identifying where one is in terms of both location and relationships, it is

helpful to formulate appropriate, realistic, short- and long-term goals. For example, if you are currently in separate locations with completely separate treatment plans and little communication (A1 on the worksheet), the most realistic short-term goal might be to increase communication (A2–A4). Once relationships are better established, it might be reasonable to consider taking steps toward co-location, and ultimately, full integration (B4–C4). Although it is certainly possible to skip steps, it is often worthwhile to proceed slowly through the various steps so that problems can be identified and solved as a group at each step. This allows time for trusting relationships to develop that is crucial to the process. These relationships can then serve as the foundation for a more fully integrated system of care in the future.

Step 3: Identify the driving forces and the barriers to change, enlist others to help tip the balance toward change.

Working toward cultural change in an organization can be time-consuming, difficult, slow, and frustrating. There are significant forces that drive people toward change and significant barriers that drive people to resist change. When the barriers and resistance to change are greater than the driving forces, change is unlikely to occur. On the other hand, when the driving forces outweigh the barriers and resistance is lowered, the opportunity exists for change to occur. In this third step, it is important to examine the driving forces and the barriers to change for each stakeholder. Once ‘‘champions’’ emerge, the task becomes to help those individuals enlist others to decrease the barriers and point out the enhanced benefits of integrated care.

Step 4: Collaboratively implement a ‘‘pilot program,’’ evaluate it, redesign it, and test it again.

Even when the balance begins to tip toward change, there may still be a significant amount of resistance. Under these circumstances, it may be worth suggesting a ‘‘pilot program.’’ Stakeholders often see this as less of a risk because it is seen as a temporary arrangement. Nevertheless, respect for everyone involved in the change (e.g., clinicians, students, secretaries, nurses, office staff, business managers, etc.) is absolutely critical at this stage. If the entire group designs the ‘‘pilot program,’’ participants can agree to convene regularly to evaluate and make changes to the program. In this way, clinical, financial, and administrative aspects of the program are all treated as equally important, each person’s opinions and thoughts matter, and each person can play an important role in helping to identify and solve problems that arise. Again, the stage is thus set for trusting relationships to develop, and that is essential to the process. In the next section, we illustrate how we used these steps to develop an integrated

primary care program in Rhode Island.

Case Illustration: Linking Systems in Rhode Island

When members of the Department of Behavioral Medicine at The Miriam Hospital (an academic medical center affiliated with Brown University in Providence, Rhode Island) began to track referrals, they discovered that a large percentage of patients

273Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

who were referred by their physicians never called to schedule an appointment. Informal discussions with patients revealed that they resented being referred for any services they perceived as ‘‘mental health,’’ and they preferred to talk to their ‘‘doctor’’ about their mental/behavioral health needs. Informal discussions with the physicians revealed that they were frustrated because they had neither the time nor the skills to help patients with all their mental/behavioral problems. Concurrently, the clinical psychology internship program in behavioral medicine at the hospital began to face new challenges. Due to changes in insurance requirements, psychology interns could see only uninsured patients, thereby diminishing the breadth of their training experience. Clinicians in behavioral medicine, referring physicians, and the Director of the Training Program were all interested in exploring new models of working together that would benefit patients, providers, and students. As described earlier in Step 1, formal and informal meetings occurred to forge relationships with organizations that wanted to be involved in change, and to identify ‘‘champions’’ within the various organizations. Very few individuals within the different groups knew anything about integrated care, so education was a key component of the process. Each time a patient was shared between a clinician in behavioral medicine, a trainee, and/or a primary care provider, it was viewed as an opportunity to discuss the merits of integrated care and explore the driving forces and the barriers to integration. As a second step, we analyzed where we were on the continuum and tried to agree

on short-term and long-term goals. At the beginning of this process, primary care offices and behavioral medicine offices were housed in different buildings several miles from each other, occasionally exchanged information, and had completely separate treatment plans (corresponding to A2 in Figure 1). The training program was located in The Department of Behavioral Medicine, but had no connection to the physician group. We set a specific goal of trying to move each group toward more regular exchange of information and coordination of treatment plans (A4). This allowed time for more trusting relationships to develop, which proved to be critical in the process. During this phase, more and more levels of health care providers were involved including nurses, technicians, office managers, and billing staffs of the different organizations. As Peek and his colleagues have emphasized (Patterson et al., 2002; Peek & Heinrich, 1995), programs that do not give ample and equal attention to the clinical, financial, and administrative spheres are doomed to fail. As described in Step 3 earlier, we identified several driving forces that provided the

impetus for the development of an integrated primary care program in Rhode Island. Because of financial changes in the health care system, primary care practices that were once housed within the hospital moved out into the community. Resources that were previously easily accessed within the same building and the same system became difficult to access. In addition, PCPs increasingly became the ‘‘gatekeepers’’ for the vast majority of their patients, requiring more time and effort. Concurrently, the average time of a medical office visit was declining due to pressures related to insurance reimbursement. To complicate matters further, psychology interns could no longer be reimbursed for treating individual patients. These changes in the health care system created a variety of problems for both

medical and behavioral health providers. Evidence had accumulated showing that behavioral health interventions can be effective, reduce health care costs, and improve patient care (Blount, 2003), but accessing such programs was a challenge, and reimbursement was an even greater challenge. The new system created a

274 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp

situation where physicians had an overwhelming number of patients who needed behavioral health treatment they could not provide, and qualified behavioral health professionals were unable to provide treatments that had been shown to be clinically beneficial and cost-effective. Although there was interest in integrating behavioral health, trainees, and primary

care, there was also a significant number of barriers. First, even finding time to discuss development of an integrated care program was an enormous challenge in itself. Discussions often occurred sporadically in hallways and cafeterias with ideas and plans written out on napkins in restaurants. Second, once barriers were identified, they seemed overwhelming and included a significant number of administrative, clinical, and financial issues. Administrative issues related primarily to staffing demands and space problems. As is true in most PCP offices, the secretarial staff was working at maximum capacity and reluctant to take on any additional administrative tasks. There was little office space available for the behavioral health providers to see patients at the primary care site. Clinical issues centered on confidentiality and record keeping. It was unclear if behavioral health visits should be documented in the medical chart, and, if so, how they should be documented. Who should have access to which sections of the chart and under what circumstances should certain sections be released to others? How would the behavioral health specialist(s) be paid? How could patients be charged without incurring additional administrative costs? How could students be utilized more effectively? At times, these issues seemed insurmountable, and it took almost 2 years of formal

and informal discussions before we were ready to launch a pilot program. The collaborators initially had very different views on how each of these issues should be handled, but in time, the group agreed on how to resolve the administrative and clinical barriers. As is commonly the case, the most difficult issues to resolve were financial. In the past, the primary care practitioners had rented space to other health care professionals, so they suggested renting space to the Department of Behavioral Medicine. The Department of Behavioral Medicine, however, could not afford to pay rent given the low rates of reimbursement for psychologists’ services. The Department of Behavioral Medicine’s view was that the PCPs should consider paying Behavioral Medicine to deliver co-located services because their providers would have to spend time traveling, behavioral health services would serve to leverage physician time, and integrating behavioral health services would increase patient satisfaction in their practice. What served as the ‘‘tipping point’’ to get us to Step 4 allowing us to launch a pilot

program? It was clear from the start that financial barriers were the primary obstacles. The tipping point came when it was decided that during the pilot phase, no money would be exchanged between the Department of Behavioral Medicine and the PCPs, and that this arrangement would be reevaluated in 6 months. The behavioral health specialists did not pay to rent space in the PCP office, and the PCPs did not pay to have the behavioral health specialists see patients in their offices. It was agreed that the behavioral health providers and PCPs would both continue billing exactly as they had done before; the only difference was that the behavioral health specialists evaluated and treated a certain percentage of patients in the PCPs office rather than in their own office. The group was well aware that it was unrealistic (clinically, financially, and

administratively) to achieve full integration from the start (e.g., to have a full-time behavioral health specialist in the PCP office). As a result, the short-term goal

275Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

became to improve communication, develop a pilot program of integrated care on a small scale (that could grow over time), and increase the number of patients who followed up for behavioral health treatment after being referred. Increasing overall access to behavioral health care was a goal for the entire group. As mentioned earlier, finding time for communication and planning is

always a tremendous challenge when developing and implementing new programs. It requires that those involved be personally committed because providers are rarely if ever reimbursed for spending time talking with one another. Thinking creatively while building on the strengths of the system that were already in place helped. In our case, communication often occurred in hallways, restaurants, and hospital cafeterias, through notes placed in charts or on chairs and desks, and through phone calls and e-mail. We all learned quickly that keeping communication brief and targeted reduced the burden of time that is in short supply for all providers. In our specific pilot program, a trainee and his or her supervisor arrived at the

PCP office once each week during the lunch break to allow time for communication with the staff. Conversations usually occurred informally around the lunch table and focused on both shared patients and ‘‘problem’’ patients in the office. Patients were then seen by the behavioral health consultant for approximately 30 minutes each. After each visit, the behavioral health consultant provided verbal feedback to the PCP (if available) and wrote a brief ‘‘Behavioral Health Consultation Note.’’ Patients signed a written consent to have these notes contained within the correspondence section of their medical chart. This section was chosen to ensure that the note would not be released unless a separate Release of Information form was signed by the patient.

Results of the Pilot Program

This pilot program began with three general questions:

* Would PCPs refer their patients for behavioral health services, and, if so, would patients come? What types of problems would trigger referrals?

* What types of interventions would most commonly be implemented? How successful would the interventions be?

* Would the service prove to be financially viable?

Which Patients Were Referred and Why?

We began by scheduling only two patients each week. Primary care physicians did not think that two patient slots would be ample; and they encouraged us to allocate more slots. Our goal, however, was to start small and build very slowly as we developed trusting relationships with the physicians, nurses, and office staff. Interestingly, it surprised everyone to find that patient slots were often empty the first month of the program. When meetings were held to discuss possible reasons, the PCPs indicated that they routinely ‘‘just forgot’’ to make such referrals. Some informal meetings were held to discuss these two areas of concern, and one of the nurses suggested putting a sign in each exam room informing patients of the available behavioral health services. This sign encouraged patients to talk with their doctor if they wanted help with problems such as weight, sleep, stress, anxiety, pain, depression, alcohol/drug use, or smoking. When patients mentioned it to their PCPs, the PCPs made the appropriate referral. After the signs went up in exam rooms,

276 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp

empty appointment slots were extremely rare, and we slowly increased the number of appointment slots as needed. We explored the characteristics of the first 68 patients to receive integrated

services, and found that they were seen for a total of 89 visits. Figure 2 illustrates the percentage of patients seen for each type of problem. As shown in the figure, the most common problem was stress and anxiety, followed by depression, pain and headache, and lifestyle factors.

What Types of Interventions Were Implemented and How Successful Were the Interventions?

The first session with the behavioral health specialist focused specifically on the issue(s) identified in the referral from the PCP. Primary goals of the first interview were as follows:

* To assess the problem from the patient’s perspective. * To assess readiness to change. * To educate the patient regarding the nature of the problem, the types of treatment

available, and what each type of treatment would entail. * To triage the patient (if appropriate) to a behavioral or mental health specialist in

the community matched to the patient’s needs as well as their insurance coverage and/or arrange a follow-up visit at the PCP office.

* To provide useful feedback to the PCP and follow-up with the PCP regarding patient progress.

Most of the patients seen (78%) were seen for only one visit in the PCP office. Of the patients seen more than once in the PCP office, almost all had been referred for either stress/anxiety or a chronic pain problem. Of the 68 patients seen, 34 (50%) were referred to mental and/or behavioral health programs within our health care system, and 8 (12%) were referred to programs outside of our health care system. Interventions for patients who were not referred to specialized programs consisted most often of motivational interviewing, education, and brief, cognitive–behavioral interventions. A major goal of the program was to increase the probability that patients

would receive appropriate care for behavioral health problems. Given this, it was of interest to explore the likelihood that patients actually followed through with

Figure 2. The percentage of patients seen for each type of problem addressed. The most common problem was stress and anxiety, followed by depression, pain and headache, and lifestyle factors.

277Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

referral suggestions. Of the 50% of patients referred to programs within our own health care system, 85% scheduled and attended their first appointment. Most of these patients were referred for treatment of depression or anxiety. Patients referred for smoking cessation were the least likely to follow through with suggestions. Of the five patients referred for help with smoking, none showed up for their next appointment. How successful were the interventions? Most patients began behavioral health

consultations with statements such as: ‘‘I’ve never talked to anyone like you before— I wouldn’t have come but my doctor really thought I should. I hope you don’t think I am crazy.’’ Clearly, these were not patients ready to accept a referral to a ‘‘mental health’’ provider on the other side of the city. Meetings held in the PCP office provided an opportunity to use the biopsychosocial model to explore problems with patients in a way they could grasp it, at a time they were ready to hear it. These visits served as a bridge to help patients link their physical and mental health, and their doctors were readily available to reinforce that message. For the PCPs, having behavioral health specialists in their office helped them (and

their staff) become more efficient. When they (or their patients) had behavioral health questions, they knew where to go for help. For the behavioral health specialists, working closely with PCPs meant better care for patients with medical problems because the care was integrated. The physician and the behavioral health specialist were working toward the same goals and could coordinate their treatment plans. This approach proved particularly important for pain patients and other patients on multiple medications that included psychotropics. Finally, for the trainees involved, the experience offered them a chance to evaluate and treat a wide variety of problems in a busy medical office, and, equally important, the opportunity to learn how to interact with nurses and physicians as colleagues. This opportunity seems particularly valuable given the vast cultural differences that exist between these groups and the ever-increasing likelihood that students will be working in integrated care settings in the future.

Was the Program Financially Viable?

When all providers are employed by one system such as an HMO, studies have shown that integrated care can at times be financially viable because of a medical cost offset. When all the providers are in different systems, however, and these systems are based on a fee-for-service model, can integrated care be financially viable? In the pilot program, the behavioral health specialist (who was, in this case, a

psychologist) visited the PCP office each week with the trainee, resulting in lost income due to travel time. The psychologist billed for services provided in the PCPs office and collected the same revenue that would have been collected had these patients been seen in her own office. We know that 50–90% of patients referred for mental health treatment do not follow through (Bloch, 1993; Glenn, 1987); and about one third of patients who make appointments do not show for their first appointments. Of the 34 patients seen in the PCP office who were referred for treatment within our own system, 29 (85%) showed up for their first visit. Given this, it seems reasonable to assume that the revenues generated amounted to more than the investment involved in traveling to the PCP office. This does not account for other factors that also impacted revenues including the increase in direct referrals from the PCP office which resulted from our collaborative efforts.

278 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp

Integrating Behavioral Health and Primary Care: Is It Worth the Effort?

Integrating behavioral health and primary care presents extraordinary challenges as well as extraordinary opportunities for those in primary care and behavioral health. The purpose of this article is to add to the growing body of evidence showing that the important question is no longer whether to integrate care, but how to integrate care. With dedication, commitment, hard work, and creativity, systems can be linked together to form networks that ultimately provide better patient care. It is our hope that the strategies and program described here will prove helpful to others attempting to build programs of integrated care.

References

Bloch, D.A. (1993). The ‘‘full-service’’ model: An immodest proposal. Family Systems

Medicine, 11, 1–7.

Blount, A. (1998). Introduction to integrated primary care. In A. Blount (Ed.), Integrated

primary care (pp. 1–43). New York: W.W. Norton.

Blount, A. (2003). Integrated primary care: Organizing the evidence. Families, Systems &

Health, 21, 121–134.

Cummings, N.A., & Follette, W.T. (1968). Psychiatric services and medical utilization in a

prepaid health plan setting. Part 2. Medical Care, 6, 31–41.

Cummings, N.A., & VandenBos, G.R. (1981). The twenty years Kaiser-Permanente

experience with psychotherapy and medical utilization: Implications for national health

policy and national health insurance. Health Policy Quarterly, 1, 159–175.

Druss, G., Rohrbaugh, R., Levinson, C., & Rosenheck, R. (2001). Integrated medical care for

patients with serious psychiatric illness: A randomized trial. Archives of General

Psychiatry, 58, 861–868.

Elhai, J.D., Richardson J., & Pedlar D.J. (2007). Predictors of general medical and

psychological treatment use among a national sample of peacekeeping veterans with health

problems. Journal of Anxiety Disorders, 21, 580–589.

Engel, G.L. (1977). The need for a new medical model: A challenge for biomedicine. Science,

196, 129–136.

Engel, G.L. (1980). The clinical application of the biopsychosocial model. American Journal

of Psychiatry, 137, 535–544.

Follette, W.T., & Cummings, N.A. (1967). Psychiatric services and medical utilization in a

prepaid health plan setting. Medical Care, 5, 25–35.

Glenn, M.L. (1987). Collaborative health care: A family oriented approach. New York:

Praeger.

Hedrick, S.C., Chaney, E.F., Felker B., Liu, C.F., Hasenberg, N., Heagerty, et al. (2003).

Effectiveness of collaborative care depression treatment in Veterans’ Affairs primary care.

Journal of General Internal Medicine, 18, 9–16.

Holder, H.D., & Blose, J.O. (1986). Alcoholism treatment and total health care utilization and

costs. A four-year longitudinal analysis of federal employees. JAMA, 256, 1456–1460.

Katon, W., & Roy-Byrne, P.P. (1989). Panic disorder in the medically ill. Journal of Clinical

Psychiatry, 50(8), 299–302.

Katon, W., & Schulberg, H. (1992). Epidemiology of depression in primary care. General

Hospital Psychiatry, 14, 237–247.

Katon, W., Von Korff, M., Lin, E., Lipscomb, P., Russo, J., Wagner, E., et al. (1990).

Distressed high utilizers of medical care. DSM-III-R diagnoses and treatment needs.

General Hospital Psychiatry, 12, 355–362.

279Developing an Integrated Primary Care Practice

Journal of Clinical Psychology DOI: 10.1002/jclp

Kessler, R., & Stafford, D. (Eds.). (2008). Collaborative medicine case studies: Evidence in

practice. New York: Springer Press.

Kroenke, K., & Mangelsdorff, A.D. (1989). Common symptoms in ambulatory care:

Incidence, evaluation, therapy, and outcome. American Journal of Medicine, 86, 262–266.

Patterson, J., Peek, C.J., Heinrich, R.L., Bischoff, R.J., & Scherger, J. (2002). Mental health

professionals in medical settings: A primer. New York: W.W. Norton.

Peek, C.J., & Heinrich, R.L. (1995). Building a collaborative healthcare organization: From

idea to invention to innovation. Family Systems Medicine, 13, 327–342.

Regier, D.A., Narrow, W.E., Rae, D.S., Manderscheid, R.W., Locke, B.Z., & Goodwin, F.K.

(1993). The de facto US mental and addictive disorders service system. Epidemiologic

catchment area prospective 1-year prevalence rates of disorders and services. Archives of

General Psychiatry, 50(2), 85–94.

Runyan, C.N., Fonseca, V.P., Meyer, J.G., Oordt, M.S., & Talcott, G.W. (2003). A novel

approach for mental health disease management: The Air Force Medical Service’s

interdisciplinary model. Disease Management, 6, 179–188.

Selden, D.R. (1997). Integration of primary care and behavioral health. In J.D. Haber &

G.E. Mitchell (Eds.), The driving forces in primary care meets mental health (pp. 13–35).

Tiburon, CA: Centralink.

Smith Jr G.R., (1994). The course of somatization and its effects on utilization of health care

resources. Psychosomatics, 35, 263–267.

Wells, K.B., Golding, J.M., & Burnam, M.A. (1988). Psychiatric disorder in a sample of the

general population with and without chronic medical conditions. American Journal of

Psychiatry, 145, 976–981.

280 Journal of Clinical Psychology, March 2009

Journal of Clinical Psychology DOI: 10.1002/jclp