|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
From Silos To Bridges: Meeting The General Health Care Needs Of Adults With Severe Mental Illnesses
Despite their higher burden of chronic general medical disorders, adults with severe mental illnesses have poorer access to and quality of general health care compared to people without mental illnesses. Because a key contributor to this situation is the separation between the mental health care sector (patients health care home) and the general health care sector, bridges to connect these systems are needed. We discuss obstacles to clinical integration and propose strategies to address them.
ALTHOUGH A LARGE PORTION OF CARE FOR PEOPLE with mental and substance use disorders occurs in the general health sector in the United States, the mental health specialty sector is the health care home for adults with severe mental illnesses.1 The poor integration between these two health systems poses major challenges to effectively meeting the complex health care needs of this population. In this paper we call attention to this problem and propose recommendations to improve the integration of mental and general health care for people with severe mental illnesses. For the purposes of this discussion, the term severe mental illness refers to schizophrenia and other psychoses, bipolar disorder, or severe depression compounded by persistent functional impairment. Prevalence of general medical conditions among adults with severe mental illnesses. Compared to people without mental disorders, adults with severe mental illnesses have higher rates of chronic general medical conditions, including hypertension, HIV/AIDS, and diabetes; a higher frequency of multiple general medical conditions; and a higher rate of premature mortality resulting from these conditions.2 Further, general medical comorbidity has a detrimental effect on their mental health, compounding already high levels of functional impairment.3 Because general medical conditions in this population are frequently missed, and many of the published rates are based on recorded diagnoses, true prevalence rates are likely to be higher than reported.4 Several factors have been implicated in the excess burden of general medical conditions among adults with severe mental illnesses. These include illness factors; socioeconomic disadvantage; substance abuse comorbidity; medication side effects; and unhealthy behavior and neglect of self-care.5 Inadequacies in the health care system, covered in greater detail below, also contribute. Is "care as usual" meeting the general health care needs of this population? Evidence suggests that people with severe mental illnesses have difficulties obtaining general health care. Not only are they less likely to have insurance than people without serious mental illnesses, but they also perceive more barriers to obtaining general health care services and report that providers of such care often dismiss their somatic complaints.6 Further, several studies indicate that the quality of general health care is poorer for people with severe mental illnesses than for less impaired people.7 For example, a study of Medicare beneficiaries found that the increased rates of mortality in this population following hospitalization for acute myocardial infarction (AMI) were partially explained by differences in quality of care.8 Several studies have been conducted in the Veterans Health Administration (VHA) system, which provides an opportunity to evaluate potential disparities in a national setting with apparent equal access. It is therefore worth noting that these studies have also documented quality gaps in preventive services, diabetes care, and care for cardiovascular conditions.9 People with severe mental illnesses who are older and who abuse substancestwo subpopulations with particularly high rates of general medical comorbidityappear to be at even higher risk for substandard quality of general health care.10 Although the observed disparities in access to and quality of care are likely to be determined by a variety of factors, the lack of clinical integration in the care of people with severe mental illnesses plays a key role.11 The Institute of Medicine (IOM) recently laid out a comprehensive strategy to improve the quality of care for people with mental and substance abuse disorders, and achieving greater clinical integration between service providers is a centerpiece of this strategy.12 Provisions in the 1963 community mental health center (CMHC) legislation requiring that physical examinations be offered on site demonstrate that the need to better address the general health care needs of mentally ill people through integrating strategies has long been recognized.13 However, recent developmentsincluding the above-mentioned IOM report and reports by the Presidents New Freedom Commission on Mental Health and the Bazelon Center for Mental Health Lawsuggest that the time is ripe to revisit this issue.14 These reports have also contributed to the discussion and recommendations that follow to effect progress in integrated care.
Our discussion of integrated care for people with severe mental illnesses pertains to Stephen Shortells construct of clinical integration, defined as "the extent to which patient care services are coordinated across people, functions, activities, and sites over time."15 Because of the centrality of care coordination to this conceptualization, the terms have been used interchangeably, as we do in this discussion. Critical prerequisites to clinical integration are communication (sharing of needed information among providers); collaboration (trusting interaction among providers, defined by shared understanding of goals and roles, effective communication, and shared decision making); comprehensiveness (effectively meeting all of the health care needs of patients); and continuity of care (timely and uninterrupted delivery of appropriate services across providers and over time).16 Organizational integration. Because clinical integration can be facilitated or hindered by nonclinical factors, we distinguish two other forms of integration here: organizational and financial. Organizational integration corresponds to the availability and functionality of linking structures that enable and sustain clinical integration. Its dimensions are structural integration, or the linking mechanisms that enable and promote clinical integration; functional integration, or the functionality of these mechanisms; and accountability, or clarity about who is responsible for the delivery of which service and where to target performance monitoring and accreditation standards.17 Linking mechanisms to bridge the mental/general health divide include formal agreements with general practitioners with expertise in treating people with severe mental illnesses; integrated clinical information systems; case management services; and deployment of general practitioners at mental health programs (called "co-location").18 Fully functional linking mechanisms can be regarded as a continuum of models of care coordination that culminates in clinical integration.19 Financial integration. Financial integrationthe degree to which financial incentives for the two care systems are aligned in the service of integrated carecan be achieved through contractual arrangements wherein risk is shared or integrated functionality is a part of the performance agreement.20 Although these forms of integration facilitate clinical integration, they are not prerequisitesin other words, much progress can be achieved along the continuum of clinical integration even when organizational and financial integration are in short supply.
In its landmark Crossing the Quality Chasm report, the IOM described the U.S. health care system as "a highly fragmented delivery system that largely lacks even rudimentary clinical information capabilities" and highlighted the poor outcomes that result from lack of coordination.21 Despite broad recognition of the need to promote the widespread adoption of interoperable information systems that support access by multiple providers to longitudinal patient information, progress has been slow and uneven.22 If successful, a recently launched public-private initiative that aims to develop a national health information infrastructure (NHII) will reap its benefits only in the next decade.23 As a result, the larger health care system remains poorly interconnected, and the care provided to the 125 million chronically ill Americans remains poorly integrated.24 However, the greater separation between general and mental health care compared with other specialty medical sectors and the inadequacy of existing linking mechanisms make coordinated health care even less likely for people with severe mental illnesses than for other populations. Mental health service delivery. Because the mental health sector is the health care home for the majority of people with severe mental illnesses, mental health care providers are uniquely positioned to improve their patients general health care outcomes. However, psychiatrists and nurses working with such patients do not routinely provide preventive counseling or screen for general health problems; nor do they monitor for adverse general health effects of medications frequently used in this population.25 Moreover, key informationincluding contact information and critical elements of the general health planis rarely known or recorded. Further, linking mechanisms that, like co-location of providers and case management, can facilitate clinically integrated care either do not exist or are functionally ineffective for the purposes of better coordination of care. For example, although mental health case management has been widely adopted by public mental health systems around the country, the predominant model is largely concerned with preventing psychiatric hospitalization.26 The small fraction of people with severe mental illnesses served by Assertive Community Treatment (ACT) programsevidence-based interventions that deliver integrated mental health and social services through multidisciplinary case-manager teams that include nursesare more fortunate in this regard.27 However, unlike the Chronic Care Model, where nurse case managers deliver an evidence-based intervention that explicitly aims to improve patients self-care and care coordination, the ACT model does not take full advantage of nurses to help improve the general health outcomes of severely mentally ill patients.28 Further, despite limited but compelling evidence of the beneficial effects of co-location on access, quality of care, and outcomes, few such programs exist in the real world.29 Other organizational supports that facilitate coordinated careincluding structured opportunities for interdisciplinary treatment planning and time free of other clinical duties so that providers may actively engage in such collaborative activitiesare infrequent in practices serving people with severe mental illnesses. Information exchange. As discussed above, the health care system has failed to fully embrace information technology (IT) and its potential for improving safety, efficiency, and care coordination. The mental health sector has lagged further behind in this regard: Most people with severe mental illnesses are cared for in practices whose information systems are underdeveloped or poorly integrated with general health practices.30 Moreover, despite existing public-sector efforts to develop an information infrastructure to support care for mental conditions, information and policy needs specific to people with these disorders have not been well addressed by the nascent NHII.31 Information exchange is also constrained by a complicated array of privacy rules and overzealous attitudes by mental health care providers. Although the Health Insurance Portability and Accountability Act (HIPAA) of 1996 generally allows providers to exchange patient information other than psychotherapy notes without requiring patients consent, it might be superseded by more-stringent federal and state laws and regulations.32 Furthermore, mental health practices might have even more restrictive policies, and psychiatrists might eschew discussing clinical information with general practitioners as a result of the strong emphasis on confidentiality during their professional training. Health care workforce. Inadequacies in the mental health workforce persist despite past efforts to improve the education and training of this heterogeneous provider group.33 The disciplines involved in the care of people with severe mental illnesses require varying depth and breadth of education and training as well as varying acceptance of and familiarity with the medical model. Further, the insufficient attention paid by all disciplines to developing and sustaining the skills necessary for interdisciplinary practice at the professional, postgraduate, and continuing education stages renders mental health practitioners ill prepared to deliver clinically integrated care.34 Lastly, licensing and credentialing assessment of mental health care providers competencies are not standardized across disciplines, states, or health plans and rarely incorporate expectations in areas relevant to general health or care coordination. It is noteworthy that expectations for mental health care competencies among general practitioners are also inconsistent with the provision of high-quality care of people with severe mental illnesses.35 Financing of care. The critical role played by the public sector in the financing of both mental and general health care for people with severe mental illnesses sets this population apart from the majority of nonelderly Americans whose health insurance is privately financed. Public-sector purchasing of care for this population occurs through insurance schemesmainly Medicaid, the federal and state program that covers three out of four people with severe mental illnessesand through direct purchasing by regional governments.36 Medicaid. Upon embracing managed care, state agencies also moved to "carve out" mental health care from the benefit package by delegating the management and financing of mental health services to managed behavioral health care organizations (MBHOs). Arrangements where benefits are not separated out are commonly referred to as carve-in contracts. Because most managed care states have either payer carve-out contracts with MBHOs (that is, public agencies separate mental and general health care in their procurement processes) or health plan subcontracts with MBHOs, most people with severe mental illnesses who are covered by Medicaid have their care funded through separate plans or through traditional indemnity programs.37 Although coordination is comparably low for carve-out and carve-in arrangements, the greater separation of mental and general health care inherent in carve-outs poses additional challenges to coordinated care for people with severe mental illnesses.38 These include lack of reimbursement for the provision of general health services by mental health care providers and vice versa (that is, cross-care), greater constraints to provider communication, and unclear accountability. Inadequacies in the procurement process have repercussions for quality of care in both kinds of contract arrangements. Because expectations for coordinated care are rarely built into requests for proposals and contracts, performance standards, monitoring capacity, and willingness to enforce penalties are typically absent from the process.39 Further, capitation rates are typically insufficient to finance adequate chronic illness care.40 Thus, health plans have few incentives to coordinate care, and in the case of separate plans, incentives could impede clinical integration.41 Incentives are also misaligned for traditional indemnity programs, as demonstrated by the lack of reimbursement for activities such as interdisciplinary meetings and case management. Direct purchasing of mental health care. State and local governments finance mental health care for the uninsured through grants awarded to local agencies. Because this financing method lacks built-in incentives for clinical integration, yet volume of services is rewarded, uninsured people with severe mental illnesses are unlikely to receive coordinated care.42 Lastly, although a growing number of federally funded community health centers offer "primary mental health care" in addition to general health care, centers clients with severe mental illnesses are unlikely to benefit from this inverse form of co-location, given the severity of their mental health care needs.43 Quality oversight. The National Committee for Quality Assurance (NCQA) and the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) monitor quality of care at the health plan and practice levels, respectively. However, these accrediting bodies have largely failed to promote coordinated care for people with severe mental illnesses. The accreditation infrastructure has few standards pertaining to the mental/general health interface, and existing standards lack specificity and "real teeth." Moreover, the NCQA does not hold all health plans to the same accreditation standards, because it assesses coordination-related performance only for carve-in health plans, exempting MBHOs from this requirement. Because standards are vague regarding responsibilities for patient care when patients belong to a general health plan and an MBHO, accountability for a frequent scenario among people with severe mental illnesses is inadequate. Oversight activities are further encumbered by the lack of performance measures suitable for assessing care processes and outcomes related to coordination of mental and general health care.
Reorganization of mental health service delivery. The goal of this reorganization is for mental health care providers and practices to assume responsibility for all of the health care needs of their patients with severe mental illnesses. Effective communication and collaboration between mental health care and general practitioners are central to this strategy. As a result, care processes would be better aligned with the goal of clinical integration. Among other improvements, psychiatrists and nurses would screen for prevalent general health problems and, in collaboration with general practitioners, would help patients with severe mental illnesses develop self-management skills.44 Consistent with best practices for chronic illness management, mental health practices would be equipped with organizational supports and case management programs redesigned to better address patients general health care needs.45 Radical changes to mental health service delivery require sustained leadership and a coherent policy agenda. Until such an agenda is fully delineated, and given the critical role of MBHOs in the care of people with severe mental illnesses, it is reasonable to entrust these entities with the responsibility of promoting the proposed changes. Promotion of patient-sanctioned communication and collaboration among providers. Attaining the goal of improved information exchange will require that relevant federal agencies ensure that the NHII addresses mental health care as fully as it does general health care. Further, federal and state governments and purchasers need to provide incentives for the adoption of clinical information systems by mental health care providers. Strategies for the latter include grants for the development of model software that can support integrated information systems and purchasing decisions that take into account the quality and extent of use of IT to facilitate coordinated care. Until such a system is fully operational, both sets of providers should use common forms of communication to seek consent and relay clinical information. Easing regulatory constraints to information exchange will require the revision of "privacy laws, regulations, and administrative practices that create inappropriate barriers to the communication of information," whereas overzealous attitudes by mental health care providers might be modified through practice-level training and written standards.46 Preparation of the health care workforce to provide coordinated care. Several improvements are needed for both types of providers to adequately address the general health care needs of patients with severe mental illnesses. While psychiatrists and nurses need to acquire competencies to screen for general health problems and help patients become more involved in self-care for all of their conditions, general practitioners need to familiarize themselves with the phenomenology, management, and prognosis of severe mental illnesses. Both sets of providers need to learn skills for interdisciplinary and collaborative work. Achieving these objectives will require short- and long-term strategies at multiple levels. Short-term progress will require continuing education initiatives and changes in licensing and credentialing requirements that could be promoted through accrediting incentives to health plans and practices. Long-term progress might be achieved by reorienting providers professional and postgraduate education toward the acquisition of knowledge, competencies, and attitudes necessary for coordinated care of people with severe mental illnessesby, for example, providing a substantial amount of clinical education in settings that can model interdisciplinary teamwork.47 Changes in provider licensing and certification requirements as well as financial and accrediting incentives to training institutions will be necessary to promote this shift. Elimination of policies and practices that offer no incentives for or discourage integrated care. Because of purchasers extraordinary power, this is a critical component of the policy agenda. For carve-out contracts, policies are needed to attenuate the negative impact of the separation between mental and general health care associated with this arrangement. Examples include reimbursement of cross-care and building shared responsibility for patients with severe mental illnesses into the procurement and accountability processes. Regardless of the contract, these processes need to include expectations for information sharing and coordination-related performance. As an example, state Medicaid agencies could request that bidders provide information on health plans coordinating practices. Further, capitation rates need to be adjusted to facilitate coordinated care. Traditional indemnity programs need to reimburse collaborative work to provide incentives for clinically integrated care. Lastly, state and local governments need to modify their grant-based approach to financing care for uninsured people so that incentives can be used to promote care coordination. Strengthening of the accreditation process. Accreditation needs to be standardized so that all health plans seeking accreditation have the same coordination-related requirements. Moreover, standards for accountability pertaining to care coordination need to be built into the process and consistently applied. To improve quality monitoring, the Agency for Healthcare Research and Quality (AHRQ) may be asked to coordinate the development of measures and standards to assess structures, processes, and outcomes associated with coordinated care for people with severe mental illnesses. One such performance measure assesses mental health care providers on their performance regarding diabetes care (testing for the HbA1C glucose level in severely mentally ill patients with diabetes). Until better measures are developed, publicly funded health plans and practices serving people with severe mental illnesses should adhere to current performance standards, and data on their performance regarding coordinated care should be made publicly available. Development of federally sponsored coordination research and demonstrations. Research is needed to evaluate the feasibility and effectiveness of (1) quality improvement interventions that, like the Chronic Care Model, provide organizational supports for coordination; and (2) adaptations of evidence-based practices for people with severe mental illnesses that might be particularly well suited to improving this populations general health outcomes. Candidates for such adaptations are ACT and illness management and recovery, an intervention that helps people with severe mental illnesses develop coping strategies and become more actively involved in their mental health care.48 Further, federally funded demonstration programs could be used to support the reorganization of mental health service delivery aimed at increasing care coordination in this population. DESPITE THE BENEFITS ASSOCIATED WITH clinically integrated care for people with severe mental illnesses, skepticism exists on the feasibility of making progress in the face of centuries of mind-body dualism and the separation between mental and general health care. In addition, concerns exist that efforts to integrate these systems might lead to loss of autonomy, resources, and specialization, hurting the very people this strategy aims to serve. Evidence indicating that carve-in plans shift funds from mental health to general health care points to potential unintended consequences of a method to organize the financing of services that may appear integrating yet, as above, has not been found to outperform carve-outs in terms of coordination.49 These findings, along with evidence that carve-outs benefit people with severe mental illnesses as a result of their better control of adverse selection and specialization of administrative and clinical expertise, belie the notion that persons with comorbid mental and general health problems are better served by carve-ins.50 To our knowledge, there is no evidence that coordinated care for people with severe mental illnesses has deleterious effects, yet it certainly has the potential for greatly improving their general health outcomes.
Marcela Horvitz-Lennon is an assistant professor of psychiatry at the University of Pittsburgh School of Medicine and Western Psychiatric Institute and Clinic in Pittsburgh, Pennsylvania. Amy Kilbourne is an assistant professor of medicine and psychiatry at the Veterans Affairs (VA) Pittsburgh Center for Health Equity Research and Promotion and the University of Pittsburgh. Harold Alan Pincus (pincusha{at}upmc.edu) is a professor and executive vice chairman of the Department of Psychiatry at the University of Pittsburgh School of Medicine. Effective 1 June 2006, he will become vice chairman of the Department of Psychiatry at Columbia University in New York City and director, Quality and Outcomes Research, at New York Presbyterian Hospital. The authors acknowledge grant support from the Robert Wood Johnson Foundation (Grant no. 48254); the National Institutes of Health (Grant no. MH52247); and the Veterans Health Administration Health Services Research and Development Service: Center for Health Equity Research and Promotion, and Health Services Research and Development Program (Grant no. MRP02269). They are also grateful to Julie Donohue for comments on an earlier draft, the editors of Health Affairs, and the anonymous reviewers for their helpful comments.
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||