Transforming Mental Health Care at the Interface With General Medicine: Report for the Presidents Commission

The President's New Freedom Commission on Mental Health was charged by President George W. Bush on April 29, 2002. Twenty-two commissioners were asked to conduct a comprehensive study of the U.S. mental health service delivery system, advise the President on methods of improving the system, identify model programs that could be disseminated to diverse settings, and develop policy options that promote the integration of effective treatments into clinical practice, improve service coordination, and improve community integration for individuals with mental illness (1,2,3).

This paper is based on a background report that was prepared to support the deliberations and recommendations of the Subcommittee on Mental Health Interface With General Medicine, which is one of 15 topical subcommittees of the New Freedom Commission.

The preparation of this report was informed by Mental Health: A Report of the Surgeon General and its supplement Mental Health: Culture, Race, and Ethnicity (4,5), a selective review of published literature on mental health services at the interface between mental health and general medicine (based on a search of MEDLINE and PsycINFO), and consultation with experts in the field. The paper is not a systematic or evidence-based literature review, such as a Cochrane review. Rather, it is a focused summary of the literature by a group of experts that was intended to help the Subcommittee on Mental Health Interface With General Medicine make recommendations in four specific areas: financing of care, performance standards, technical assistance, and provider training. The original literature review was updated in late 2004 to add relevant citations from 2002 to 2004.

Findings and recommendations from the subcommittee were reported to the full Commission in Arlington, Virginia, on January 8, 2003 (6,7).

Mental illness and the medical care system

Several decades of research have documented the high prevalence of common mental disorders in general medical settings. Evidence suggests that community-dwelling people with common mental disorders frequently present in general medical settings, such as primary care clinics, during the course of an illness episode, although relatively few receive specialty mental health care (8,9). Unmet needs for mental health care are particularly prevalent among older adults, children and adolescents, individuals from ethnic minority groups, uninsured patients or patients with low incomes who are seen in the public sector, and individuals with mental health problems who present with primarily physical symptoms (9,10,11,12,13,14,15,16,17,18).

Overall, about half of the care for common mental disorders is delivered in general medical settings, leading Regier and colleagues (8) to describe general medical settings as the "de facto mental health care system" in the United States. Thus general medical settings represent an important—perhaps the single most important—point of contact between patients with mental disorders and the health care system. The role of primary care providers is even more important for older adults (19) and patients from racial and ethnic minority groups. For example, among Mexican Americans, almost two-thirds of consultations for treatment of a mood or anxiety disorder are with primary care providers (20). Similarly, African Americans with mental health concerns are more likely to see a primary care physician than a psychiatrist (21).

Efficacious treatments exist for most common mental disorders (4), as do treatment guidelines for primary care providers and mental health specialty care providers for prevalent conditions, such as depression. Nevertheless, a majority of patients with mental illness are not treated effectively in primary care settings or elsewhere (4,8,9)

Co-occurring mental and medical illness

In addition to increasing the personal and societal burdens of mental illness, low rates of effective mental health treatment are also likely to increase the burdens associated with physical illness. Mental disorders frequently co-occur with other medical disorders. For instance, adults with common medical disorders have high rates of depression and anxiety (22). Major depression among patients with chronic medical illnesses increases the burden of somatic symptoms (23), causes additional functional impairment (24,25), and increases medical costs (26,27). Depression also impairs self-care and adherence to treatments for chronic medical illnesses (22) and causes increased mortality, which is perhaps best documented among patients after they experience a heart attack (28). Some of this excess mortality may be explained by direct biological effects of mental disorders on the cardiovascular system (for example, neurohumoral dysregulation and platelet changes) or behavioral factors (for example, poor diet, smoking, alcohol abuse, and obesity). However, poor quality of preventive, acute, and chronic medical care may also contribute to premature cardiovascular deaths (29).

Barriers to effective care

Patient, provider, and system factors all contribute to poor quality of care at the interface of mental health with general medicine. For instance, patients may not recognize or correctly identify their symptoms; be reluctant to seek care, particularly because of stigma; or be reluctant to adhere to treatment recommendations. Primary care providers may lack the necessary training and confidence to provide appropriate treatment for mental health problems, and even well-trained and motivated primary care providers are limited in what they can accomplish in a 12- to 15-minute office visit during which they may also be addressing multiple medical and social problems. Finally, mental health benefits are typically more restricted and heavily managed than other medical benefits, and few insurance plans cover providers' costs for implementing screening, care management, and other proactive services that studies have shown to be effective in increasing the rate of evidence-based treatment, as we discuss below.

Similar factors underlie the poor quality of care for mental disorders other than depression, for the mental health problems of patients with co-occurring medical conditions, and for the medical conditions of people with mental illness. The personal and societal consequences of these quality gaps at the interface between mental health and general medicine can be substantial—unnecessary suffering, functional impairment, mortality, economic losses, and health care costs.

Quality improvement strategies

Numerous studies have assessed strategies to improve care at the interface of general medicine and mental health. Much of this work has been done in the area of depression, because it is one of the most common disorders seen in general medical settings, and because efficacious treatments have existed for several decades. More recently, parallel research has examined strategies to improve care for anxiety and somatoform disorders and for severe and persistent mental illness among patients with comorbid medical disorders. In this section, we summarize evidence on such strategies.

Case finding and diagnosis

Much of the early research on improving care for mental disorders in general medical settings has focused on case finding and diagnosis. Numerous studies since the 1970s have focused on developing screening instruments for common mental disorders in primary care (4,30), such as the PHQ primary care study (31).

Several studies have coupled screening with systematic feedback of depression diagnoses to primary care providers Although some of these screening studies showed an increase in the rate of diagnosis of mental disorders in primary care, no screening studies have demonstrated a significant effect on clinical outcomes of mental disorders, such as depression (32,33). This evidence suggests that screening and provider feedback are necessary but not sufficient to improve outcomes of adults with common mental disorders.

Correspondingly, the U.S. Preventive Services Task Force recently recommended that primary care providers screen adult patients for depression, as long as they have systems in place to ensure accurate diagnosis, effective treatment, and follow-up (34).

Referral to specialty mental health care

One option for addressing mental health problems in general medical settings is to increase rates of referral to mental health specialty providers, but relatively little research has addressed how to improve these referral processes. Research has shown that primary care providers perceive themselves as having much worse access to suitable mental health providers than to other medical specialists (35,36). Studies have also shown that one-third to one-half of primary care patients who receive a referral to mental health specialty care do not follow through with such a referral (37,38). Finally, successful entry into the specialty mental health care system does not guarantee effective care; many patients treated in specialty mental health clinics do not receive care consistent with existing treatment guidelines (39,40). These findings suggest that although the referral process between general medical settings and mental health specialty care should be improved, reliance on specialty referral alone will fall short of addressing currently unmet needs for mental health care at the interface between mental health and general medicine.

Treatment guidelines and provider education

In the 1990s, several organizations developed and published consensus guidelines for the treatment of common mental disorders, such as depression in primary care (41,42). These guidelines recognized that efficacious treatments, such as antidepressant medications and certain types of psychotherapy, which were originally developed and tested in specialty mental health settings, could also be delivered effectively in primary care settings. Clinical practice guidelines can be an important component of strategies to improve quality of care. However, the consensus in the field is that distributing practice guidelines to primary care providers does not by itself improve quality of care or patient outcomes. It can be difficult for providers to use such guidelines without additional help and changes in the service delivery system. Diagnosing and treating depression can also be complicated by the fact that as many as 75 percent of depressed primary care patients present predominantly with somatic symptoms. Despite the wide availability of practice guidelines, only about one-third of internal medicine residents in a recent survey of primary care training programs reported that they felt "very prepared" to diagnose and treat depression in the outpatient setting, and residents' level of comfort regarding the management of depression was far lower than for eight other common conditions of outpatients (43).

In this context, researchers have tested strategies that combine practice guidelines with comprehensive training of providers, with mixed results. For instance, one study showed that a comprehensive 20-hour training program improved physicians' knowledge and depression outcomes at three months but not at one year (44). Another study found that intensive guideline-based training of providers on diagnosis and treatment of depression increased the rate of appropriate diagnosis only slightly (from 36 percent to 39 percent) and had no effect on patients' depression status relative to usual care (45).

Overall, these findings suggest that provider education programs, including continuing medical education, academic detailing, and continuous quality improvement, may be an important component of comprehensive quality improvement strategies and that they are likely necessary but not sufficient by themselves to improve care for common mental disorders (46,47)

Collaborative care

Motivated by evidence that strategies such as screening, dissemination of practice guidelines, provider training, and referral to mental health specialists do not appear to be effective on their own, researchers have developed programs that aim to improve care at the interface of mental health and general medicine by training psychiatrists, psychologists, and other mental health specialists to provide depression care in collaboration with patients' primary care providers or by training members of the primary care team, such as nurses, in additional skills to care for patients with mental health needs. Such professionals are often supported by consultation from a designated mental health professional. We broadly refer to such interdisciplinary care management programs as "collaborative care" (48,49).

Although the programs vary, most effective treatment programs have two key elements. The first is systematic care management by a nurse, social worker, psychologist, or other trained clinical staff to facilitate case identification, coordinate an initial treatment plan and patient education, provide close follow-up, monitor progress, and modify treatment if necessary. Care management can be provided in a primary care clinic or by telephone. The second key element is consultation between a care manager, a primary care provider, and a consulting psychiatrist or other appropriate mental health specialist.

Several recent trials of collaborative care models have attempted to maximize the cost-effectiveness of collaborative care by using a process described as "stepped care" (49,50,51,52,53). Stepped care usually begins with relatively low-intensity interventions, such as antidepressant medications prescribed by a primary care provider and care management provided by telephone or in the primary care clinic. Under the supervision of a consulting psychiatrist or other appropriate mental health specialist, patients who are not helped by such initial treatments are shifted to progressively more intensive treatment approaches, including referral to specialty mental health care as needed.

Effective treatment programs recognize that usual primary care is better suited to address acute, time-limited medical problems rather than chronic illnesses that require ongoing monitoring and management (54,55). With respect to depression, for instance, half of patients do not improve adequately with the first course of an antidepressant medication, yet the time and resource constraints of usual general medical practice provide little opportunity to monitor patients' progress and adjust treatment.

More than ten studies of collaborative care models for depression in a wide range of health care systems have demonstrated that they are more effective than usual care (56,57,58,59,60). Such models have been shown to improve clinical outcomes, employment rates (61,62), functioning, and quality of life, and they are cost-effective compared with other commonly used medical interventions (61,62,63,64,65).

Collaborative care appears to be particularly beneficial for individuals from ethnic minority groups, who traditionally have low rates of appropriate care, and this approach may reduce ethnic disparities in quality of care (63,66,67). These models have also been successfully extended to adolescents (68) and to older adults, who often have comorbid medical disorders (52,69,70,71). Analyses from the two largest collaborative care trials for depression suggest that these programs are equally effective for patients with and without comorbid medical disorders (72,73). Similar collaborative care models have been developed for anxiety disorders in primary care (74,75).

Suggestions for high-quality care for common mental disorders at the interface of general medicine and mental health are summarized in the box on this page (17,48,49,54,76,77,78,79,80,81,82).

Improving medical care for patients with severe mental illness

Several studies have involved population-based screening for medical problems of patients with mental disorders (83,84). These screening programs did not involve specific efforts to address identified health problems, nor did the researchers assess the impact of the screening per se on clinical outcomes. Nevertheless, there is little reason to expect that screening and case identification alone will be effective in improving treatment of medical conditions among people with severe and persistent mental illness.

To improve medical care for patients with severe mental disorders, some academic leaders have called for systematic efforts to train psychiatrists to recognize and treat their patients' basic medical problems (85,86). As with programs focusing on mental health training for primary care providers, this approach emphasizes knowledge deficits among providers but may not adequately address time and financial constraints, as well as patient factors.

Others have advocated and developed dual training programs in which physicians train and become board eligible in psychiatry and a primary care specialty (87). Currently, about 40 such training programs exist (88). Evidence about the effects of these programs is limited, but the results of one study suggest that only a minority of dually trained physicians actually practice both specialties (87). It is also unlikely that such programs could ever train sufficient numbers of physicians to meet the needs of all individuals with comorbid medical and mental disorders.

A collaborative care framework similar to the models developed in primary care in which mental health providers and primary care practitioners collaborate in the care of patients, may hold promise for improving medical care for patients treated in specialty mental health settings. Although substantially less research has been done in this area than in the treatment of depression in primary care, the results of one randomized trial suggest that on-site, integrated delivery of primary medical care can improve quality and outcomes of medical care for patients with serious mental disorders (89).

Barriers to disseminating quality improvement

With the availability of effective quality improvement models, particularly for depression, efforts are being made to develop strategies to disseminate such models (90,91,92). Results from this research suggest that successful programs are multifaceted and include various combinations of physician and patient education, "tools" to diagnose and monitor common mental disorders in general medical settings, patient registries or tracking systems, care management, and more effective collaboration between primary care and mental health providers (49,76,77,78,7993,94,95). More generally, however, a range of organizational and financial systems affect the scope and quality of health care at the interface of mental health and general medicine.

One major barrier is access to the mental health system. Insurance benefits for mental health care are generally more limited than for general medical care, and some patients, particularly those with severe mental illness, quickly exhaust their benefits. In concept, mental health parity laws are intended to eliminate such differences, and full parity is required in some states and for federal employees. However, although parity laws are an important step, these laws are unlikely in and of themselves to be sufficient to ensure equal access to care (96).

A second barrier is that health insurance reimbursement mechanisms often do not provide incentives for use of evidence-based models of care for mental disorders or for many other chronic conditions (79,97). For instance, neither public nor private health insurance plans provide adequate reimbursement for care management or specialty consultation services, which are key components of collaborative care models for depression and other chronic conditions.

The lack of reimbursement inhibits dissemination, because providers will implement collaborative care interventions only if the benefits of doing so exceed the costs. Although collaborative care and other evidence-based interventions are highly cost-effective from a societal perspective, many of these benefits are experienced as "externalities"—that is, outside the mental health system (for example, via improved medical care) or outside the health system (for example, via improved employment outcomes, reduced need for social services, or reduced suffering of patients' families, friends, and coworkers). Overcoming these potential means of market failure will require purchasers, such as employers and governments, to use targeted financial incentives to promote the use of evidence-based services by providers (97,98).

Other barriers may arise from the separation of mental health care from the mainstream of medicine by managed mental health care (4). Managed behavioral health care organizations and pharmacy benefit managers have become the prevailing form of providing insurance coverage for mental health treatment and prescription drugs, respectively. The separation or carving out of these benefits from the general medical plan can create important organizational barriers to primary care patients who need mental health specialty services or to the effective collaboration of primary care providers and mental health specialists. Several managed behavioral health care organizations have recently made efforts to address this barrier by providing better integrated disease management in collaboration with primary care, and the National Committee for Quality Assurance has released a set of performance indicators for managed behavioral health care organizations that provide incentives for effective collaboration with primary care.

Finally, uninsured patients and individuals covered under Medicaid and other public-sector programs for low-income populations have particularly high rates of common mental disorders, such as depression and anxiety. Although individuals with low incomes who have severe and persistent mental illnesses and who are covered under Medicaid often receive services in specialty mental health care settings, such as community mental health centers, those with common mental disorders, such as depression or anxiety, have limited access to specialty mental health care but substantial unmet medical, emotional, and social needs. For these reasons, it is particularly important to support models of care in which primary care providers caring for patients with low incomes can be supported by specialty mental health professionals. One example of such an effort is an initiative to improve care for depression sponsored by the Bureau of Primary Health Care. More research is needed on the effectiveness of this and similar efforts.

Opportunities to improve quality of care and policy

Recommendations

We have described considerable deficits in the quality of care for people with mental disorders, quality improvement models that have been shown to be effective in addressing some of these deficits, and current barriers to the widespread adoption of such models. We recognize that many of the issues described in this report could benefit from additional research, but we conclude with recommendations that could mitigate some of the existing barriers and facilitate the use of evidence-based quality improvement models. The recommendations are listed in the box on page 42.

Financing for collaborative care services. Our first recommendation is that public and private insurance programs explicitly cover evidence-based collaborative care for common mental disorders. Many of the services in evidence-based collaborative care models, such as in-person consultations and psychotherapy, are reimbursable for providers under current financing mechanisms. Similarly, psychotropic medications are generally covered by health insurance programs, now including Medicare. However, as we have described, key components of the collaborative care model are not currently reimbursable, particularly care management services that provide proactive follow-up and coordination of care for chronic mental disorders and mental health consultations to primary care-based practitioners or care managers that do not have to involve face-to-face contact with patients. Coverage for these services should be coupled with appropriate oversight to ensure that they are provided appropriately, using evidence-based treatment protocols.

One major target for reform must be care management activities, particularly those falling outside the scope of traditional outpatient office visits. Care managers should be reimbursed for time spent consulting with primary care and specialty mental health providers and for coordinating care by these providers, for patient education activities, and for providing proactive monitoring and follow-up (including via telephone). We note that research suggests that people with various types of training, including nurses, clinical social workers, psychologists, and others, can provide these services effectively if they have adequate training and supervision (49,50,5192,99). Some of these providers are not currently reimbursed to provide such care management in some health plans, particularly Medicare. Mental health consultation may involve in-person consultations in which a mental health specialist, such as a psychiatrist or psychologist, evaluates the patient in a medical or mental health setting. Mental health consultation may also involve consultations with a treating primary care provider or a primary care-based care manager that do not involve direct patient contact. Such consultations can occur in person or by telephone and should be documented by the consultant, the care manager (if applicable), and the primary care provider to facilitate coordination of care.

In insurance programs that reimburse providers under a fee-for-service arrangement, such as traditional Medicare, the most direct way to cover these care management and consultation services would be to make them billable in their own right. Mechanically, such an approach may require the creation of new procedure codes, or the adoption of existing codes. It may be as simple as removing existing regulatory barriers to collaborative care, such as current Federally Qualified Health Centers restrictions by Medicare and Medicaid on reimbursing primary care and mental health visits at a single clinic on the same day. In plans with capitation-based or other types of contracts with providers, collaborative care services would become part of the benefit to which patients are entitled, and plans and providers would incorporate these new benefits into rate negotiations.

A second policy recommendation related to the financing of care includes a recommendation that the government achieve better coordination of the funding and the clinical care provided to clients of publicly funded community clinics for medical, mental, and substance use disorders, such as community health centers funded by the Health Resources and Services Administration (HRSA) and community mental health centers supported by the Substance Abuse and Mental Health Services Administration (SAMHSA). Improved coordination of funding streams and clinical care by federal, state, and local government agencies and providers could improve the effectiveness of care at the interface of mental health and general medicine, reduce duplication of services, and prevent clients with mental disorders from "falling through the cracks."

Finally, we endorse a recent recommendation by the Institute of Medicine (100) that the federal government develop research and demonstration programs that study financial incentives to improve the quality of care. These should include incentives for health plans to coordinate medical and mental health services or to meet target outcome criteria for common mental disorders, such as depression. In cases in which behavioral health care is carved out, financial incentives should be shared between managed care organizations, managed behavioral health care organizations, and, when applicable, pharmacy benefit managers. For such financial incentives to be effective, clear performance standards for care at the interface of medicine and mental health need to be developed and applied by federal, state, and private payers.

Performance standards. Our second recommendation concerns performance standards. Successful implementation of evidence-based quality improvement can be strengthened by increasing the accountability of health care providers through performance measurement. Although practice guidelines and performance standards have been developed for a number of common mental disorders by various government agencies and professional organizations, these standards are usually developed for application in either primary care or specialty mental health care settings, and they generally do not address care at the interface of general medicine and mental health and the collaboration of general medical and mental health practitioners.

We recommend that federal and state agencies, private insurers, and accrediting organizations, such as the National Committee for Quality Assurance and the Joint Commission on Accreditation of Healthcare Organizations, develop clear performance standards for the care of individuals with mental disorders. Such standards should extend existing guidelines to include performance standards for care at the interface between mental health and general medicine and mechanisms to audit and track adherence to performance standards. The new standards should not only include measures of the process of care but also focus on achieving desired health outcomes. Examples of such outcome-oriented criteria include the percentage of depressed patients in a health plan who are documented to experience at least a 50 percent reduction in depression symptoms as measured by a standard instrument, such as the Patient Health Questionnaire (PHQ-9) (101) over a six-month period and the proportion of patients who do not reach 50 percent improvement who receive a change in treatment, such as augmentation of medications, addition of psychotherapy to medications, or referral to specialty mental health care.

We recommend that similar performance standards be developed for the recognition and care of common medical disorders among individuals with severe mental illnesses who are treated in specialty mental health care settings. Medical conditions and treatments should be included along with psychiatric and psychosocial issues in patients' problem lists and treatment plans. If mental health care settings cannot provide basic medical evaluation and treatment on site, they should arrange for appropriate medical care and coordination of care with qualified primary care providers.

Technical assistance. Third, we recommend that public and private stakeholders develop technical assistance programs to help health care providers implement evidence-based models of care at the interface of general medicine and mental health. Relevant organizations include government agencies such as the Agency for Healthcare Quality, the National Institute of Mental Health, SAMHSA, HRSA, the Department of Veterans Affairs (VA), the Centers for Medicare and Medicaid Services (CMS), large insurers, and provider organizations

Successful implementation and dissemination of evidence-based quality improvement programs at the interface of mental health and general medicine will require more than well-educated providers. Organizations that wish to implement and sustain evidence-based models of care at the interface of mental health and general medicine could greatly benefit from technical assistance programs that could provide the necessary materials and "tools," technical expertise, and consultation to help them implement, disseminate, evaluate, and bill for such programs. Effective technical assistance programs could support provider training, sponsor local and national quality improvement processes, support the development and standardization of necessary information technology, develop business models for sustainability, and develop and disseminate performance standards to monitor care at the interface of mental health and general medicine. Several private and public organizations have developed toolkits to help improve treatment of mental disorders in general medical settings; such efforts, although intended for somewhat different target audiences, can be highly complementary and should be supported to ensure widespread dissemination of evidence-based care models (102,103,104).

Provider training. Finally, we recommend that public and private stakeholders develop strategies to improve the training of medical and mental health practitioners in the care of patients at the interface of mental health and general medicine and that further research be conducted to evaluate the effectiveness of such programs.

Relevant stakeholders include government agencies that fund or support medical education, such as CMS through Medicare and Medicaid, the VA, and HRSA; professional schools and graduate training programs in medicine, nursing, social work, and clinical psychology; professional organizations, such as the American Medical Association, the American Psychiatric Association, and the American Psychological Association; and licensing, credentialing, and accrediting bodies, such as the American Association of Medical Colleges, the Liaison Committee on Medical Education, and the Accreditation Council for Graduate Medical Education (ACGME). These organizations all have important opportunities to affect the training and certification of primary care and mental health care practitioners in ways that can improve collaboration.

Training programs for medical students and primary care providers should cover the fundamental knowledge and skills required to diagnose and treat common mental disorders, such as depression and anxiety disorders, and to make effective referrals to mental health providers for patients who do not respond to first-line treatments in general medical settings or who prefer treatment from a mental health specialist. Treatment of patients with mental disorders and serious medical disorders often requires a process of interpersonal and emotional growth; several approaches have been developed to teach psychosocial aspects of care to primary care providers (105), and some primary care programs have integrated support groups, such as Balint groups, although there is limited evidence about the effectiveness of such groups.

Training programs for mental health providers should include training in brief, structured psychotherapies that can be delivered in medical settings; effective consultation to and collaboration with primary care providers; recognition of common medical disorders; and effective coordination of the medical care for these disorders with primary care practitioners. Mental health and primary care providers should also receive training in evidence-based behavioral techniques to support effective self-management of chronic medical disorders. A number of reimbursable procedure codes already exist that can be used by psychologists and other qualified mental health providers to bill for counseling individuals with chronic medical disorders. To improve providers' knowledge and skills in these areas, curricula in training programs for nurses, physicians, social workers, and psychologists should be revised.

Such curricular policies have been variably implemented in training requirements for primary care physicians and mental health specialists by the appropriate specialty boards, but additional research is needed to examine their effectiveness. It is also important to point out that although provider knowledge and training may be necessary to improve mental health care, they may be only a small part of the solution. It is likely that system changes in the way care is delivered will also be needed.

Besides making curricular improvements, it is important to change the location of training in mental health. Too often, mental health providers are still trained in "professional silos," such as psychiatric hospitals or clinics, with little exposure to patients with common mental disorders in general medical settings and with little opportunity to learn effective consultation to and collaboration with primary care practitioners. Coyne and associates (106), among others, have pointed out that psychologists working in primary care can make a number of important contributions to the care of patients with common mental disorders, including diagnosis and treatment of emotional disorders as well as related activities in the areas of health promotion and pain management.

Sample training programs at the interface of mental health and general medicine include the VA program that trains psychiatrists or psychologists to work as part of a multidisciplinary team in primary care (107), a program that trains medical students and psychiatric residents in primary care settings (85), and training programs in "psychosomatic medicine" or "primary care psychiatry" for primary care physicians, psychiatrists, and psychologists. The recent formalization of psychosomatic medicine as an accredited subspecialty of psychiatry with ACGME-accredited fellowship training programs may support the development of mental health practitioners with greater skills in interfacing with medicine. Training programs that emphasize training at the interface between mental health and general medicine should be expanded with support from the federal government, and more research is needed to evaluate their effectiveness.

The role of government

Much of the challenge in improving mental health care is the fact that both the locus of and the responsibility for that care is so diffuse. Persons with mental disorders may be treated in specialty mental health settings or general medical settings, and care may be financed by the federal government, state government, or private payers. The fragmentation that creates barriers to high-quality care may itself raise challenges to implementing the funding changes, performance standards, technical assistance, and provider training that are needed to overcome those barriers. What, then, is the most appropriate site at which to begin targeting efforts at quality improvement?

As previously discussed, a major barrier to implementation of evidence-based care is the fact that health plans and private payers typically reap only a portion of the benefits that derive from improving quality. Because government represents a broader range of interests, it can play a critical role in spearheading these efforts. A recent report by the Institute of Medicine (100) points out that the federal government is in a unique position to assume a strong leadership role in driving the health care sector to improve the safety and quality of health care services that are provided to the approximately 100 million beneficiaries of the six major government health care programs—Medicare, Medicaid, the State Children's Health Insurance Program, TRICARE, the VA health system, and the Indian Health System—and that the federal government should "vigorously pursue purchasing strategies that encourage the adoption of best practices through the release of public-domain comparative quality data and the provision of financial and other rewards to providers to achieve high levels of quality."

Medicare is currently sponsoring several large-scale demonstration projects that seek to improve care for common chronic conditions. It will be important for these efforts to include mental disorders, both as primary conditions and as comorbid disorders affecting care for other chronic diseases. More broadly, we recommend that the federal government develop a set of sample purchasing specifications for contracts that are based on current best practices at the interface of medicine and mental health services as well as effective methods to measure performance at this interface.

Conclusions

The Subcommittee on Mental Health Interface With General Medicine (6) outlined several policy options in the areas of financing of services, performance standards, technical assistance, and provider training (see box on page 42). In its final report the President's New Freedom Commission on Mental Health (108) prioritized the following recommendations with regard to improving mental health care at the interface with general medicine:

• Recommendation 4.4: Screen for mental disorders in primary health care, across the life span, and connect to treatment and supports.

• The Commission suggests that collaborative care models should be widely implemented in primary health care settings and reimbursed by public and private insurers.

• Medicare, Medicaid, the Department of Veterans Affairs, and other federal- and state-sponsored health insurance programs and private insurers should identify and consider payment for core components of evidence-based collaborative care, including case management, disease management, supervision of case managers, and consultations to primary care providers by qualified mental health specialists that do not involve face-to-face contact with clients.

The Commission also recommended that the federal government "could better coordinate the funding and the clinical care provided by publicly funded community health clinics to consumers with multiple conditions, including physical, mental, and co-occurring substance use disorders. This effort would include improved coordination of care between HRSA-funded community health clinics and SAMHSA- or state-supported community mental health centers."

We agree with the emphasis on financing evidence-based treatment programs at the interface of mental health and general medicine, and we believe that financing of such services should be closely tied to outcome-based performance standards for care of common mental disorders at this interface.

Acknowledgments

This article is based on a report prepared for the Subcommittee on Mental Health Interface With General Medicine of the President's New Freedom Commission on Mental Health. The content of this article does not imply endorsement by the U.S. Government or by the Commission. The authors assume full responsibility for the accuracy of the content. This work was funded under task order 280-99-1003 of contract 280-99-1000 with the Substance Abuse and Mental Health Services Administration. The authors acknowledge input from Henry T. Harbin, M.D., the chair of the Subcommittee on Mental Health Interface With General Medicine and other subcommittee members, including Anil G. Godbole, M.D.; Norwood W. Knight-Richardson, M.D., M.B.A.; Nancy C. Speck, Ph.D.; and Deanna F. Yates, Ph.D. The authors are also grateful for comments by Kenneth B. Wells, M.D., M.P.H.; Jeanne Miranda, Ph.D.; Gregory Simon, M.D., M.P.H.; Elizabeth Lin, M.D., M.P.H.; Michael Von Korff, Sc.D.; Christopher Callahan, M.D.; Herbert C. Schulberg, Ph.D.; John Williams, M.D., M.H.S.; and David A. Pollack, M.D.

Dr. Unützer and Dr. Katon are affiliated with the department of psychiatry and behavioral sciences at the University of Washington Medical Center, Box 356560, Seattle, Washington 98195 (e-mail, [email protected]). Dr. Schoenbaum is with RAND in Arlington, Virginia. Dr. Druss is with the School of Public Health at Emory University in Atlanta.