Research on Community-Based Mental Health Services for Children and Adolescents, Volume 14 (Research in Community and Mental Health)

LIST OF CONTRIBUTORS Steven M. Banks The Bristol Observatory, Bristol, VT, USA Julia Bell University of New England ...

Author: William H. Fisher

36 downloads 1849 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

LIST OF CONTRIBUTORS Steven M. Banks

The Bristol Observatory, Bristol, VT, USA

Julia Bell

University of New England College of Osteopathic Medicine, Biddeford, ME, USA

Leonard Bickman

Center for Evaluation & Program Improvement, Peabody College, Vanderbilt University, Nashville, TN, USA

Kathleen Biebel

Center for Mental Health Services Research, Department of Psychiatry, University of Massachusetts Medical School, Worcester, MA, USA

Laura Compian

Child Services Research Group, Department of Psychiatry, University of California, San Francisco, CA, USA

Maryann Davis

Center for Mental Health Services Research, Department of Psychiatry,University of Massachusetts Medical School,Worcester, MA, USA

William H. Fisher

Center for Mental Health Services Research, Department of Psychiatry, University of Massachusetts Medical School, Worcester, MA, USA

Jeffrey L. Geller

Department of Psychiatry, University of Massachusetts Medical School, Worcester, MA, USA vii

viii

LIST OF CONTRIBUTORS

Rebecca Grusky

School of Policy, Planning and Development, University of Southern California, Los Angeles, CA, USA

James P. Guevara

Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

Sarah McCue Horwitz

Departments of Epidemiology & Biostatistics, Psychiatry & Pediatrics, Case Western Reserve University School of Medicine, Cleveland, OH, USA

Nancy Koroloff

Regional Research Institute for Human Services, Graduate School of Social Work, Portland State University, Portland, OR, USA

David S. Mandell

Center for Mental Health Policy and Services Research, Department of Psychiatry, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

Denine Northrup

Department of Psychology, Western New England College, Springﬁeld, MA, USA

John A. Pandiani

The Bristol Observatory, Bristol, VT, USA

Susmita Pati

Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

Melissa Pearrow

Department of Counseling and School Psychology, University of Massachusetts, Boston, MA, USA

Tracy J. Pinkard

Center for Evaluation and Program Improvement, Vanderbilt University, Nashville, TN, USA

List of Contributors

ix

Abram Rosenblatt

Child Services Group, Department of Psychiatry, University of California, San Francisco, CA, USA

Christine VanVleck

The Bristol Observatory, Bristol, VT, USA

Peter Whelley

Moultonborough School District, Moultonborough, NH, USA

INTRODUCTION: CHILDREN, ADOLESCENTS, AND MENTAL HEALTH SERVICES RESEARCH: AN OVERVIEW OF EMERGING PERSPECTIVE William H. Fisher As an undergraduate psychology major, I attended a university in which the psychology department was tilted heavily toward issues of child development. But in my child psychology class, my professor began the course by placing what we know about child development in a historical context. One of our ﬁrst (and most enjoyable) assignments was to visit our local art museum. Our mission was speciﬁc: Look at the faces of the children in paintings from the 16th, 17th, and 18th centuries, and report back. What did we see? We agreed that in at least some of those paintings, children’s faces were distinctly ‘‘un-childlike.’’ In fact, the children in those photographs often looked like, and were even dressed like, small adults. What was the purpose of assigning this exercise in art appreciation in a child psychology class? It was to make an important point: that our understanding of children and of the phenomenon of childhood itself, have evolved signiﬁcantly since those times. The notion that childhood is a distinct developmental stage that the needs of children are unique, the minds of

Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 1–9 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14001-5

1

2

WILLIAM H. FISHER

children are not the same as the minds of adults, all are products of the early 20th century and largely due to the efforts of early psychologists such as G. Stanley Hall (1905). Had my young child psychology professor (who, incidentally, was now a faculty member in the department that 60 years earlier Hall had helped to found) coordinated this line of argument with what my sociology professor was discussing at the same time, something he called the ‘‘sociology of knowledge,’’ this discussion would have been framed in the context of the ‘‘social construction’’ of childhood – the notion that our understanding of phenomena is socially derived and that the actions taken by societies and individuals in any area of life are grounded in those constructions (Berger & Luckmann, 1966). This includes some of our most basic experiences – including the human life span. We experience the human life span as more or less a ‘‘seamless continuum,’’ but the developmental milestones it encompasses have signiﬁcant ramiﬁcations for the larger society. For at least a century, western behavioral science has recognized that childhood, adolescence, adulthood, and senescence are distinct and important periods in which individuals undergo continual and signiﬁcant cognitive, moral, and social development. This recognition has come to affect the way western societies address the needs of children and in the laws that govern their behavior and the ways in which adults treat them. We understand, for example, that children and adolescents do not possess the same capacity for decision making as adults, even when they may sometimes look and behave like adults. We understand that, their medical needs are different from those of adults, they may be less culpable than adults when they break the law, and they may be vulnerable in a variety of ways that must be recognized when the larger society designs interventions to manage them or even simply allows them to work. All the while, however, we retain a developmental perspective reminding us that, milestones notwithstanding, the lifespan is a continuum in which early events and experiences affect later behavior and actions. In the medical profession, the salient features of childhood and adolescence led to the growth of pediatrics as a major specialty, and to the founding of ‘‘children’s hospitals’’ which catered to the speciﬁc medical and psychosocial needs of sick children. As with all medical specialties, pediatrics evolved its own body of specialized knowledge and procedures, as well as its own complement of subspecialties in surgery, cardiology, oncology, endocrinology, etc. Pediatrics evolved its own professional organizations, journals, residency training, and other social trappings of major medical specialties. Even when, in the 1970s, there was an effort abroad in medicine to better integrate pediatrics and adult medicine within the larger

Introduction

3

framework of ‘‘family practice,’’ the knowledge base and social infrastructure of pediatrics remained intact. As with physical medicine, ‘‘discovery’’ of the unique features of childhood and adolescent behavior led ultimately to the development of subdisciplines – child psychology and psychiatry, which constructed psychopathology differently as well. As is typical of professional groups, these subspecialties have developed a specialized knowledge base (Freidson, 1970) that has taken the form of a professional construction of age-speciﬁc deviant behaviors. Deﬁned as ‘‘disorders,’’ these behaviors have been ‘‘medicalized’’ (Freidson, 1970). Codiﬁed and subsumed into the diagnostic taxonomy reﬂected in the Diagnostic and Statistical Manual’s section on ‘‘Disorders of Childhood and Adolescence’’ (American Psychiatric Association, 1994). As is also characteristic of medical subspecialties, these professional groups derive and maintain their power and unique status through their ability to apply their deﬁnitions of disease to a speciﬁc clientele (Freidson, 1970). As such, these disciplines have created and imposed unique deﬁnitional categories that are differentiated chieﬂy by age. Indeed, so pronounced are these subspecialty differences that certain behaviors that seem similar with regard to social presentation and meaning may be given different labels, largely as a function of an individual’s age. For example behavioral patterns which include marked violation of social norms may be described as ‘‘conduct disorder’’ for individuals under the age of 18, but will be called ‘‘antisocial personality disorder’’ in adults. While having had the ‘‘childhood version’’ of the disorder by age 15 is one criterion for being diagnosed with the ‘‘adult’’ disorder, and while these behavioral patterns share a status with respect to violation of social norms, they are nonetheless classiﬁed separately by mental health professionals. In the latter part of the 20th century greater recognition of the special needs of children and a general de-emphasis on institutions as solutions to social problems led to the creation of an increasingly differentiated set of public agencies designed to deal with various problems associated with childhood and adolescence. Large orphanages were replaced with a foster care system, overseen in most jurisdictions by a ‘‘department of social services.’’ Reform schools were replaced by departments of youth services, and a separate juvenile justice system was created for managing offenders under the age of 18. Indeed, much juvenile offending was given the special status of ‘‘delinquency’’ and special courts were developed for trying cases involving juveniles. In general, then, it can be said that the care and treatment of children and adolescents has gone forward in a regulatory environment that is much different from that focused on adults. Social welfare policy also views

4

WILLIAM H. FISHER

children and adults differently for a variety of purposes, important among them the determination of eligibility for beneﬁts provided under various entitlement programs. For example, separate regulations apply to women and dependent children with regard to welfare assistance and Medicaid. These differences and their ramiﬁcations have been particularly pronounced within behavioral health; children under the age of 18 have been eligible to receive Medicaid reimbursement for extended inpatient treatment in psychiatric specialty hospitals, a service that is denied to older individuals until they reach they age of 64. The co-evolution of social welfare policy and the deﬁnitional regimes of psychiatric/psychological subspecialties with regard to children and adults has led to signiﬁcant differences in the management of their psychiatric disorders. For the last quarter-century these differences in managerial and treatment approaches have induced signiﬁcant structural change in a range of organizations, including those administering mental health services. In many locales, public mental health authorities maintain separate inpatient services, residential programs, and arrays of specialized community-based services for children and adults. These differences are reﬂected in the evolving managerial structures of these agencies, which typically feature a separate managerial and service functions located within a ‘‘child mental health’’ division, which oversees the provision of services to eligible clients under that age group. This age-based differentiation in mental health service systems can be attributed to a number of factors. One, historically, has to do with the recognition that the large institutions that once housed psychiatric patients of all ages, while in some cases unﬁt for almost everyone, were particularly inappropriate for younger persons. In Massachusetts, for example, the mid1980s saw the ﬁnal push to exclude persons under the age of 18 from adult psychiatric hospitals. This exclusion, of course, necessitated the development of alternative services to meet the needs of individuals in this age group who would have been hospitalized in adult institutions. In addition, federal programs such as Medicaid generated eligibility criteria and funding streams that reinforced the separation of adult and child mental health services.

RESEARCH ON MENTAL SERVICES FOR CHILDREN AND ADOLESCENTS As we have discussed, recognition of the specialized needs of children and adolescents has taken place late in many domains. The same could be said

Introduction

5

with regard to the mental health services research enterprise. Kimberly Hoagwood, a prominent child mental health services researcher and for many years the program ofﬁcer overseeing the National Institute of Mental Health’s services research portfolio in the area of child mental health services research, recently observed that research in this area has lagged behind that for adults by roughly 10 years (Hoagwood, 2005). Indeed, as late as 1999, Burns was calling for the development of a services research agenda for youth with serious emotional disturbances (Burns, 1999). Arguably, this gap has not been the product of indifference; rather, this delay likely reﬂects the greater degree of complexity encountered when attempting to ‘‘bound’’ the set of services and service settings that focus on mental health needs of children. Service systems for adults are complex, but, as the chapters of this volume collectively suggest, not anywhere near as complex as those for children. This increased complexity is due, perhaps, to the fact that children routinely and simultaneously traverse a host of such systems and settings – schools, pediatricians’ ofﬁces, and most importantly, their homes. Mental health problems that may be apparent in one setting may not be as obvious in another, and the consequences of symptom expression in different contexts maybe very different. The chapters in this volume present a picture, albeit an incomplete one, of this complex array of settings and services, and of the potential effects and prospects of new policies shaping services and approaches to treatment. In the ﬁrst section of our volume, ‘‘Research on Services – from Birth to Young Adulthood’’ we begin with the most fundamental of relationships – the mother–child unit. Sarah Horwitz, Julia Bell, and Rebecca Grusky provide an overview of the current approach to identifying and treating postpartum depression. They argue that some of the same settings in which infants and toddlers are treated could be ones where depression in their mothers might be detected. Other settings, such as the workplace, could serve this purpose as well. Clearly, while the focus of such treatment is an adult mother, the implications for the well-being her child are critically important. But as Horwitz and her colleagues indicate, this spectrum of services and settings largely fails in this regard. One of the problems facing researchers striving to understand the scope of the mental health needs of children is the multiplicity of settings in which those needs are identiﬁed and addressed. None is more important than the school. Going to school is what most children ‘‘do.’’ For many it represents their ﬁrst ofﬁcial foray into a world in which social, functional, and emotional demands will be placed on them. Increased sensitivity to these needs, coupled with legal mandates to provide special education services, has made

6

WILLIAM H. FISHER

the school a key mental health services outpost for many children. Melissa Pearrow and Peter Whelley, a school psychologist, provide an overview of the challenges facing public schools in fulﬁlling this mission, a challenge that includes the need to interface the schools services with those of other system entities. As we noted earlier, public policy has evolved toward a system of separate entitlements for children and adults. While the appropriateness and good intentions of such a bifurcation are unassailable, there are, nonetheless, problems that arise at the boundary of childhood and adulthood. This age-driven differentiation has led not only to the development of different service provision mechanisms for children and adults, but also to different eligibility criteria for receiving those services. Put simply, children diagnosed with one of the range of ‘‘serious emotional disorders’’ found in the child psychopathology taxonomy may be eligible to receive a broad array of services from their state mental health agency until their late teens – typically until the age of 18. But, while these disorders do not ‘‘magically disappear’’ on one’s 19th birthday, eligibility for receiving services from the state mental health agency, including access to residential and other critical support provided by that agency effectively ends. This occurs because, after age 18, individuals are deemed eligible for state mental health agency services based on a different set of criteria, a protocol which usually requires that one be diagnosed with one of the major adult psychiatric disorders, such as schizophrenia. Thus, while their mental health and related problems persist, their supports effectively end. That this represents a signiﬁcant failure of policy should seem obvious, but until recently this problem had not been a focus of discourse for mental health policy or services researchers. In ‘‘The Great Divide: How Mental Health Policy Fails Young Adults,’’ authors Maryann Davis and Nancy Koroloff address this issue. Speciﬁcally, they provide data on service eligibility policies in 46 states to determine the extent to which current policies affect the continuity of services for youth in child mental health systems as they transition from adolescence to adulthood. Their policy analysis demonstrates that each state’s child mental health policy differed from its adult population policy, generally in the direction of more narrow adult criteria. In Part II, ‘‘Evaluating and Examining ‘Systems of Care,’’’ we present three chapters that examine one of the central concepts in contemporary mental health services delivery for children and adolescents – the ‘‘system of care’’ (Stroul & Friedman, 1986). If the African proverb ‘‘It takes a village to raise a child’’ is true, it would appear that it takes a multi-faceted, multisetting system to care for a child with emotional disorders. In this section of

Introduction

7

the volume we present three chapters focusing on evaluation and assessment of systems of care. David Mandell, James Guevara and Susmita Patel describe a new system entity, the ‘‘medical home,’’ and compare it with the system of care model. The paper provides a useful overview of the evolution of thought that created the current approach to providing multi-disciplinary approaches to caring for medical and psychological treatment needs of children. It also points out the need, endemic in mental health services design, for those creating the medical home model to learn the lessons obtained through the creation of the system of care model. Applying the term ‘‘system’’ to a collection of providers naturally invites the question of whether services are coordinated and actually ‘‘appear as a system’’ from those embedded within it. This question is addressed in the chapter by Denine Northrup, who applies the widely used network analysis approach to evaluate service system conﬁguration. The paper is heuristic as well as evaluative, useful describing how the evolving network analytic methodology can be used to examine critical issues in the complex of providers and agencies seeking to meet the needs of young persons with mental health needs. How large is a locale’s system of care? How many children are served? How do we measure demand for services when children can be seen in so many different settings? In their chapter, John Pandiani, Christine VanVleck, and Steven Banks offer a methodological approach to this problem using the large administrative databases maintained in most jurisdictions but rarely combined to examine this issue. Using probabilistic approaches, they show how the number of individuals served in multiple facets of the system of care can be examined without violating the stringent safeguards on data privacy recently imposed by the Health Information Privacy and Protection Act. In the third and concluding part of the volume, ‘‘Systems of Care and Evidence Based Practice: Theoretical and Conceptual Issues in Research and Policy Analysis,’’ we present three papers which explore conceptually the system of care model, the use of evidence-based practices, and their relationship. Two papers, one by Tracy Pinkard and Leonard Bickman and another by Kathleen Biebel and Jeffrey Geller, discuss how well systems of care and other conceptually driven service models fare in the new policy environment that emphasizes evidence-based practices. Rosenblatt and Campion, in the volume’s ﬁnal chapter, describe methods by which research on systems of care and research on evidence-based practice can be integrated. While these papers collectively address issues in the delivery of mental health services to children and adolescents, many of the issues they raise are not peculiar to that treatment system alone; indeed many of these issues pervade the entire mental health policy and services research arena.

8

WILLIAM H. FISHER

Together, the papers in this volume present the services research community a thoughtfully derived set of conceptual issues and systems analyses which should help in the further honing of a research agenda for examining critical areas of service delivery for children and adolescents, appropriately integrated services, and necessary continuity. Any such agenda should also recognize the importance of the ‘‘non-specialty sector’’ – providers of treatment such as schools, primary care pediatrics and the juvenile justice system, and determine these providers the other services with which their clientele come into contact. This volume has several important omissions. For example, recent studies indicate that a substantial number of the youths detained in juvenile justice settings have serious emotional and substance use disorders (Grisso & Underwood, 2003). There has been in this system an increased emphasis on screening for such disorders as well as honoring the constitutional rights of correctional detainees to medical and psychiatric care. How this treatment is integrated with that provided by the locales’ broader system of care and how youthful detainees with mental health needs are linked to those systems upon their release is but one of many key issues for future mental health services research efforts. We have not included a chapter on this topic, and clearly one should be present. If, as Kimberly Hoagwood observed, research on services for children and adolescents has lagged behind that focusing on adults, there is clearly important work ahead. As this collection of papers suggests, research in this area is fascinating for its scope, its multi-actor and multi-system features and critically important because of the precious population toward which these services are geared.

REFERENCES American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders: Fourth edition, DSM-IV. Washington, DC: American Psychiatric Association. Berger, P. L., & Luckmann, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. Garden City, KS: Doubleday. Burns, B. J. (1999). A call for a mental health services research agenda for youth with serious emotional disturbance. Mental Health Services Research, 1(1), 5–20. Freidson, E. (1970). Profession of medicine: A study in the sociology of applied knowledge. New York: Harper and Row. Grisso, T., & Underwood, L. (2003). Screening and assessing mental health and substance abuse disorders among youth in the Juvenile Justice System. In: Research and program brief, National Center for Mental Health and Juvenile Justice. Delmar, NY: Policy Research Associates.

Introduction

9

Hall, G. S. (1905). Adolescence; Its psychology and its relations to physiology, anthropology, sociology, sex, crime, religion and education. New York: Appleton. Hoagwood, K. (2005). Challenges and opportunities in child and adolescent mental health services. In: Paper presented at the 133rd annual meeting of the American Public Health Association, Philadelphia, PA, December 10, 2005. Stroul, B. A., & Friedman, R. M. (1986). A system of care for children and youth with severe emotional disturbances. Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center.

THE FAILURE OF COMMUNITY SETTINGS FOR THE IDENTIFICATION AND TREATMENT OF DEPRESSION IN WOMEN WITH YOUNG CHILDREN Sarah McCue Horwitz, Julia Bell and Rebecca Grusky BACKGROUND Depression is a prevalent, debilitating condition that will replace cancer as the second leading cause of morbidity within the next decade and, according to the Global Burden of Disease Study, ranks number one in disabilityadjusted life years for females 5 years and older worldwide (Blehar & Oren, 1997; Murray & Lopez, 1996). Depression in the workplace has been linked to increased absenteeism and productivity loss, is equal to the costs of diabetes and hypertension, and these costs are almost equal to the direct costs of depression treatment (Kessler et al., 1999; Marlowe, 2002; Druss, Rosenheck, & Sledge, 2000; Elinson, Houck, Marcus, & Pincus, 2004). A national study of individuals 15–54 years documented a lifetime prevalence of 17.1% and found that depression was more common in females, young adults, and those with less education (Blazer, Kessler, McGonagle, & Swartz, 1994; Kessler, McGonagle, Swartz, Blazer, & Nelson, 1993; Substance Abuse and Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 13–31 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14002-7

13

14

SARAH MCCUE HORWITZ ET AL.

Mental Health Services Administration, 2000; Kessler et al., 1994a, 1994b; Bebbington et al., 2003). Depression in younger women is a particularly important problem since it has the potential to affect not only women but also their children (The Florida State University Center for Prevention & Early Intervention Policy, 2002; Lovejoy, Graczyk, O’Hare, & Neuman, 2000; Goodman & Gotlib, 1999; Guttmann, Dick, & To, 2004; McLennan, Kotelchuck, & Cho, 2001; Davies, Howells, & Jenkins, 2003; Grace, Evindar, & Stewart, 2003). Estimates of depression in women with children (maternal depression) range from 10 to 42%. The majority of information on the prevalence of maternal depression comes from postpartum studies rather than studies of mothers with young children, although there are data to suggest that mothers may develop depression throughout the children’s early years (McLennan, et al., 2001). Maternal depression may, therefore, be either a continuation of depressive symptoms beyond the immediate postpartum period or the development of depressive symptoms in women with young children. Identifying and treating maternal depression is critical, since depression in mothers can have signiﬁcant effects on child development beyond the postpartum period. Grace et al. (2003) in a review of the literature on the effect of postpartum depression on child cognitive development and behavior, concluded, ‘‘chronic or recurrent maternal depression, rather than postpartum depression per se is likely related to later effects on the child’’. This conclusion is sharpened by the ﬁndings of McLennen and colleagues (2001). Using a national sample, they documented that 36% of women with elevated symptoms of depression remain highly symptomatic at 1-year post the original assessment (McLennan et al., 2001). This suggests that a signiﬁcant portion of these women may be chronically depressed and clearly are not being recognized or treated. Consequently, large numbers of children may be exposed to the deleterious effects of chronic maternal depression. While maternal depression has received little attention, the gender difference in depression for late adolescents and adults has received considerable attention, although the causal mechanisms for this are not well understood (Weissman & Klerman, 1977; Cyranowski, Frank, Young, & Shear, 2000; Hankin et al., 1998) and are likely to be multifactorial (Cyranowski et al., 2000). Recent cross sectional and longitudinal studies suggest that this gender difference is due to increased ﬁrst onsets of depression in females. These onsets begin to emerge in the 13–15 year age range but magnify in the 15- to 18-year period (Hankin et al., 1998; Wells, Sturm, Sherbourne, & Meredith, 1996) and coincide with the beginning of the childbearing years.

The Failure of Community Settings

15

Although depression in women with young children is a prevalent and important problem, the extant literature suggests that few depressed women are either identiﬁed or treated and, if they are treated, few receive adequate care (Wells et al., 1996). A lack of identiﬁcation and treatment exists despite the availability of easy to administer, reliable, valid, and acceptable screening instruments and effective therapies (Wells et al., 1996; Spitzer et al., 2000; Oklahoma State Department of Health, 1995; Herrick, 2002). National efforts, such as recommendations for screening by the US Preventive Services Task Force, a Surgeon General’s Report on Mental Health, AHCRP guidelines for the treatment of depression, a Congressional Resolution, and National Depression Screening efforts, have also failed to improve identiﬁcation and treatment outcomes for maternal depression (U.S. Preventive Services Task Force, 2002; U.S. Public Health Service, 2000; Postpartum Depression Resolution, 1999; U.S. Department of Health and Human Services, Agency for Health Care Policy and Research, 1993; NIMH DART (Depression Awareness Recognition and Treatment Program), 1995). Given the importance of depression as a problem for young women and their children and the clear evidence that the condition goes largely unidentiﬁed and untreated, we asked what was known about the identiﬁcation and treatment of depression within the service sectors commonly interacting with women with young children.

METHODS During February and March 2002, we asked females with children under the age of 10 years who were employees and visitors of one division of an academic medical center what systems and services they interacted with in a typical week. We then asked to whom they would turn for assistance with or information about a mental health problem, such as depression. In addition to family and friends, women identiﬁed eight different service sectors that they commonly interacted with, were inﬂuenced by, or would seek assistance, advice, or information from about depression: mass media, religious organizations, places of employment, child care programs, community services, primary care physicians, obstetrician/gynecologists (OB/GYN), and their children’s pediatricians (Fig. 1). To learn what these service sectors were providing regarding depression outreach and treatment, we undertook an exhaustive review of both the peer reviewed and popular literature and contacted many national organizations,

SARAH MCCUE HORWITZ ET AL.

16

Pediatric Medical Care

Employment

Adult Medical Services: Primary Care

Fig. 1.

Mass Media

Adult Medical Services: OB/GYN

Women with Children

Religious Organizations

Community Services

Child Care Programs

Model of Service Sectors that Women Seek Information from About Depression.

such as the National Head Start Association, National Employee Assistance Providers, and the Maternal and Child Health Bureau of the U.S. Department of Health and Human Services. To access the peer-reviewed literature and information on service sectors, the key words of depression, maternal depression, perinatal depression, and postpartum depression were used in association with the words screening, services, outreach, treatment, and prevention. These searches were paired with each of the service sectors. Medline and Psychinfo databases were researched as well as the Cochrane reviews. The search engines Google and Yahoo also were utilized. In all, over 400 articles written in English were reviewed.

RESULTS Mass Media Virtually everyone living in the United States is exposed to information transmitted via television or the Internet. In US households, 98% own at least one television, 51% own a computer, and 41.5% have Internet access (U.S. Dept. of Commerce Census Bureau). A number of public information programs and media campaigns on depression and mental illness have been mounted, including Depression/Awareness Recognition and Treatment (NIMH D/ART), the National Public Education Campaign on Clinical Depression (NPECCD), National Depression Screening Day and National Mental Illness Awareness Week (NIMH DART (Depression Awareness

The Failure of Community Settings

17

Recognition and Treatment Program), 1995; National Depression Screening Day; National Public Education Campaign on Clinical Depression; Greenﬁeld et al., 2000). NPECCD estimates that its efforts reached 93% of the population an average of 11 times (National Public Education Campaign on Clinical Depression). However, most of these efforts have not been formally evaluated and, consequently, their impacts have not been quantiﬁed (Greenﬁeld et al., 2000; Hirschfeld et al., 1997). We could identify no national efforts speciﬁcally targeting maternal depression. We also searched media campaigns within state departments of health and found that very few speciﬁcally recommend routine screening for maternal depression. Among those that do is the Maryland Center for Maternal and Child Health of the Maryland Department of Health and Mental Hygiene, which recently published a pamphlet on postpartum depression with an emphasis on risk factors, recognition, severity, and where to go for help. The Center was also awarded a grant from the federal Health Resources and Services Administration for a project called Women Enjoying Life Longer (WELL), designed to improve access to preventative health services, including depression screening and referral. However, the available reports on women’s health in Maryland did not speciﬁcally recommend screening for maternal depression (The Center for Maternal and Child Health, Maryland Department of Health and Mental Hygiene, and Health Care Answers, 2002). Maryland is also one of the 30 states currently participating in the CDC’s Pregnancy Risk Assessment Monitoring System (PRAMS), an ongoing survey of new mothers. Additional questions to the main PRAMS questionnaire speciﬁcally ask about symptoms and behaviors that may be indicators of maternal depression. However, most states choose not to use these additional questions. When they are used, high rates of depression are common and usually one half or less of women with symptoms report receiving treatment (Oklahoma State Department of Health, 1995; Herrick, 2002). Religious Organizations Seventy-ﬁve percent of US women report some religious afﬁliation (Kosmin & Mayer, 2001). Religious organizations have, in the past, provided successful sites for hypertension and breast cancer screenings (Kosmin & Mayer, 2001; Markens et al., 2002). Further, studies have shown that religious involvement has provided many women a form of social support and, therefore, have suggested that church programs to address the psychological needs of congregations may be particularly useful in helping

18

SARAH MCCUE HORWITZ ET AL.

women cope with depression (Mirola, 1999). However, we could identify no studies examining either identiﬁcation or referral and treatment of depression within religious organizations. Employment Almost 65% of women with children 6 years and younger work outside the home; 67% of ﬁrst-time pregnant women work, and 52% of women returned to work within 6 months of the birth of their ﬁrst child (Maternity Leave and Employment Patterns, 1961–1995). Depression in the workplace is estimated to affect 20% of US workers annually. Two-thirds of these individuals go undiagnosed and untreated. The medical disability costs for depressed workers are equal to those of diabetes and hypertension (Marlowe, 2002; Druss et al., 2000). Depression in the workplace has been linked to increased absenteeism and productivity loss. These problems result in costs almost equal to the direct costs of depression treatment (Kessler et al., 1999). Employee Assistance Programs (EAPs), which were started in response to alcohol and drug abuse problems, are increasingly being used to identify and treat depression. In 1998, there were 20,000 operating EAPs in the US. Thirty-nine percent of all private worksites with 50 or more employees had EAPs in 1998. In 2000, 62.1 million workers had access to EAPs (French et al., 1999; Reynes, 1998; Oss & Clary, 1998). The use of EAPs for treatment of mental health issues appears to be successful although little data are available. Sprang and colleagues (1992) described a six to eight session EAP-based intervention to treat depression (Sprang, 1992). Using a drop out comparison group, the difference between pre and post scores on the Beck Depression Inventory were statistically signiﬁcant for the brief structured therapy intervention group and were unchanged for the comparison group. Despite these encouraging results, we could identify no studies that looked at the treatment of maternal depression through EAPs. Child Care Fifty-one percent of children 2 years and younger are cared for by adults who are not their parents. Sixteen percent of children are in center-based programs. Although many services are recommended, particularly for children who live in families with multiple risk factors, identiﬁcation and treatment of maternal depression is rarely offered (Behrman et al., 1995).

The Failure of Community Settings

19

Head Start, currently serving over 800,000 low-income children, has historically been silent on maternal depression. Recently, however, parental mental health appears to be receiving more attention in Head Start. An evaluation of National Head Start/Public Early Childhood Transition Demonstration found that 40% of primary caregivers screened positive for depressive symptoms when their children entered Kindergarten (Ramey et al., 2000; Lanzi et al., 1999). In October 2000, the Head Start Bureau commissioned a review of the literature on maternal depression (Wessel & The Ellsworth Associates Research Team, 2000). Some Head Start Programs are screening for maternal depression and Early Head Start, which provides services to pregnant women and their families, mandates education on maternal depression (U.S. Department of Health and Human Services, Administration for Children and Families, Head Start Bureau, 2002). There are a number of recent federal efforts around depression in Head Start families. The Agency for Children, Youth and Families and the National Institute of Mental Health have developed a collaborative mental health research initiative (NIMH Research Center on Poverty, Risk & Mental Health, 2002). In March 2004, the Health and Human Services, Administration for Children and Families awarded grants to examine depression in Head Start. One of the grants to the Children’s Hospital in Boston, Massachusetts, is a partnership with Head Start and other community organizations to develop an intervention model using manual-based training and a service program to prevent, identify and treat depression in parents (HHS Awards $2.9 Million To Support 30 Head Start Innovation Projects, HHS News 25 March 2004). Community Services: Home Visiting Programs Although not universal, home visiting programs have become more numerous in the US. The major US programs aim to improve the lives of children through changes in the knowledge, attitudes, and behaviors of their parents. This is usually achieved with practical assistance that links them to other community services (Behrman, 1999). US home visiting programs vary in stafﬁng patterns, time of onset, length, intensity, and focus. Results of their evaluations show modest positive results for parents but few consistent positive results for children due to low enrollment rates, less intensity of service than originally intended and poor retention rates (20–67% of participants drop out prior to completion) (Behrman, 1999). Few home visiting programs assess or treat mothers’ mental health. Hawaii’s Healthy Start Program did assess maternal depression, but, at 1 year, found no difference

20

SARAH MCCUE HORWITZ ET AL.

in rates between the intervention and control groups (Duggan et al., 1999). In contrast to the US, home visiting programs in European and Commonwealth countries have been used successfully to identify and treat maternal depression (Parke & Hardy, 1997; Holden, Sagovsky, & Cox, 1989). Adult Medical Services: Primary Care The general medical sector is the only source of care for 40–60% of depressed individuals (Wells et al., 1996; Regier et al., 1993). Studies of primary care patients suggest that between 2.2 and 18.9% of patients may be depressed. Estimates vary by criteria used, assessment procedures, and patient population (Olfson et al., 2000). Approximately 50% of all depressed patients are not identiﬁed and few receive adequate antidepressant treatment or speciﬁc psychotherapies (Wells et al., 1996; Katon, Von Korff, & Lin, 1992, 1997; Ormel et al., 1991; Pe´rez-Stable et al., 1990). Considerable work in primary care has demonstrated that quality of care for depression can be achieved. However, underidentiﬁcation and inadequate treatment persist (Wells et al., 1999, 2000). Deﬁciencies are attributed to patient presentation of problems, unwillingness to accept treatment and noncompliance, provider attitudes and a lack of training and competing demands, practice and organizational issues such as increased expectations for productivity, poor reimbursement for psychosocial care and limited access to specialty care, and conceptual issues such as the model of psychiatric caseness in primary care (Wells et al., 1996; Good, Good, & Cleary, 1987; Eisenberg, 1992; Klinkman, 1997; Klinkman et al., 1998; Williams et al., 1999). While there is massive literature on detection and treatment of depression in primary care, there is little attention given speciﬁcally to the issue of maternal depression. Epperson (1999) published a review article on the detection and treatment of postpartum depression in Primary Care and Wisner and colleagues (2002) have suggested standards for the identiﬁcation and treatment of postpartum depression (Epperson, 1999; Wisner, Parry, & Piontek, 2002). This lack of attention to maternal depression may be due to the utilization patterns of young women. In the 2000 National Health Interview Survey, women with children 36 months and younger reported low use of outpatient medical services in the past 12 months (13% made no visits; 15% made one visit) (National Center for Health Statistics, 2000). Thus, although the US Preventive Task Force (USPSTF) recommends that adults be routinely screened for depression in clinical practices, few young women see primary care physicians regularly (U.S. Preventive Services Task Force, 2002).

The Failure of Community Settings

21

Adult Medical Services: OB/GYN Between 70 and 84% of US women receive pregnancy related services prior to the third trimester (Beck et al., 2002; Lipscomb et al., 2000). Studies in the US and other countries suggest that depression occurs throughout pregnancy and in the immediate postpartum period (Cox, Conner, & Kendell, 1982; Cox, Murray, & Chapman, 1993; Kumar, 1994; Kumar & Robson, 1984; O’Hara et al., 1990; Chaudron et al., 2001). Studies in OB/GYN settings suggest that depression is prevalent. Miranda et al. (1998) found 21.5% of women seen in GYN clinics at San Francisco General had current depression, few of whom had regular care. Spitzer et al. (2000) identiﬁed (using the Prime-MD) 20% of OB/GYN outpatients as having mental health disorders and documented that providers did not identify 77% of women with disorders. Yonkers et al. (2001) screened women visiting inner city clinics at the ﬁrst postpartum visit and found that 6.5–8.5% of these women were depressed. Georgiopoulos et al. (2001) universally screened for postpartum depression in community postnatal care sites and found that 10.7% of the population was diagnosed with postpartum depression. Several studies outside the US also suggest that depression is prevalent in OB/ GYN settings. Buekens et al. (1998) found that 19% of Belgian women visiting gynecologists scored very high on depression questions. Of those, physicians identiﬁed only 51%. Sundstrom and colleagues (2001) screened women attending two outpatient gynecology clinics in Northern Sweden. In this study, 10.1% were identiﬁed with major depressive disorder and 12.4% with minor depressive disorder. Diagnoses were largely unknown prior to this study (Sundstro¨m et al., 2001). Hsiao et al. (2002) identiﬁed as depressed 36% of women visiting a hospital outpatient gynecologic clinic in Northern Taiwan. A study of the medical charts for women delivering in Olmstead County, Minnesota in 1993 revealed that only 3.7% of women had documented in their charts at least one symptom of postpartum depression in the year following delivery (Bryan et al., 1999). A screening of women with the Edinburgh Postnatal Depression Scale who delivered within Olmstead County and who were Olmstead County residents at the 6 week postpartum visit revealed that 11.4% had elevated depression scores (Georgiopoulos et al., 1999). Lack of identiﬁcation and treatment persist despite the availability of valid, reliable, and acceptable screening and diagnostic instruments, excellent published risk beneﬁt discussion and treatment recommendations, published descriptions of the problem and the diagnostic challenges and the

22

SARAH MCCUE HORWITZ ET AL.

availability of effective pharmacologic and psychotherapeutic treatments (Wisner, Zarin, & Holmboe, 2000, 2002; Reid et al., 1998; Yonkers & Chantilis, 1995). Interventions in the antepartum period have produced mixed results, although recent efforts targeted at vulnerable women and examining intervention impacts by levels of self esteem found positive results (Zlotnick et al., 2001; Matthey et al., 2004; Hayes, Muller, & Bradley, 2000; Elliott et al., 2000; Buist, Westley, & Hill, 1999). Secondary interventions after the diagnosis of postnatal depression show consistent signiﬁcant improvement in depression (O’Hara et al., 1990; Lumley & Austin, 2001; Ray & Hodnett, 2001). One possible explanation for the lack of identiﬁcation and treatment of maternal depression may be that women are not actively seeking care. One reason is that they do not recognize their symptoms as depression and may instead attribute somatic features of depression to other physical causes. Women may also be reluctant to seek care because of the stigma associated with mental health problems (Alvidrez & Azocar, 1999; McIntosh, 1993; Whitton & Appleby, 1996). Another possible explanation is that OB/GYN providers do not screen for depression. OB/GYNs infrequently use screening instruments or formal diagnostic instruments (Williams et al., 1999; Whitton & Appleby, 1996). Eighty percent of OB/GYNs report receiving no training in the treatment of clinical depression and 60% have not completed a CME course on the topic (Schmidt et al., 1997). OB/GYNs, compared to internists and family practitioners, are also less likely to perceive a responsibility for treating depression and to prescribe antidepressant medications, and they doubt their ability to manage depression. LaRocco-Cockburn, Melville, Bell, and Katon, (2003) found that many OB/GYNs believe routine screening for depression would be difﬁcult to implement in everyday practice, and some question whether screening in OB/GYN settings improves outcomes. The study suggests that the majority of OB/GYNs believe they have a responsibility to identify depression, but they are not usually provided with the appropriate resources and training to screen for and treat depression. The importance of training OB/GYN residents to diagnose and treat maternal depression has recently received attention (Stevens & Diehl, 2003). Current policies may also be contributing to the lack of identiﬁcation and treatment. While the US Public Health Guidelines recommend screening adults for depression in primary care, there are no speciﬁc screening recommendations for maternal depression and the studies used to establish these guidelines most often include general primary care patients or focus on the elderly (U.S. Preventive Services Task Force, 2002; Pignone et al., 2002).

The Failure of Community Settings

23

Further, screening for depression is not routinely evaluated in the CDC monitoring survey for maternal and child health (PRAMS) (Cox et al., 1982). Although a recent American College of Obstetricians and Gynecologists (ACOG) Committee Opinion suggests psychosocial evaluation, no speciﬁc recommendations are made for screening for behavioral health issues in general or depression speciﬁcally (ACOG Committee Opinion No. 292, 2003). Child Health Settings Ninety-six percent of children 0–4 years have a usual source of care (Federal Interagency Forum on Child and Family Statistics, 2001). Ninety-one percent of children 36 months of age or less reported on in the 2000 National Health Interview Survey had received health care from their usual provider in the past 12 months. Sixty-ﬁve percent of these children had four or more visits in the previous 12 months (Weitzman & Auinger, 2002). Although the importance of identifying maternal depression has been discussed in the pediatric literature for many years (Zuckerman & Beardslee, 1987; Green, 1994), and well-accepted guidelines for pediatric health supervision suggest screening (Green & Palfrey, 2002), work by Heneghan and colleagues (2000) documents that pediatricians fail to recognize mothers with depressive symptoms. Twenty-one percent of mothers scored at least 30 on the PSI, and pediatricians identiﬁed only 34% of these mothers as depressed (Heneghan et al., 2000). In the routine service of Stockholm’s Well Baby Clinics, 2% of 1,128 mothers were identiﬁed as depressed by providers, as opposed to 14.5% by the Edinburgh Postnatal Depression Scale (BagedahlStrindlund & Monsen, 1998). Kahn and colleagues (1999) found that among mothers with a variety of health, mental health, and social problems who had visited pediatric clinics, 17% had no regular source of care and 39% reported two or more barriers to getting care. More than 85% of these women said they would welcome or not mind inquiries about these issues from their child’s pediatrician (Kahn et al., 1999). Reasons for this lack of identiﬁcation of maternal depression in pediatric settings include patient, provider, and systems factors. Mothers often do not recognize symptoms of depression (Seidman, 1998) and they may be reluctant to discuss these issues with their child’s pediatrician particularly if they do not have an ongoing relationship with a pediatrician (Heneghan, Mercer, & DeLeone, 2004). Forty-three percent of pediatricians do not believe it is their responsibility to detect maternal depression, and 55–69% of pediatricians are not conﬁdent in their ability to manage maternal depression.

24

SARAH MCCUE HORWITZ ET AL.

In addition, pediatricians do not use screening instruments to detect maternal depression (Olson et al., 2002). Their failure to use readily available screening instruments is due to concern about time, skills, and resources necessary to manage identiﬁed problems (Cheng et al., 1996). Lastly, policies and organizational issues may also be responsible for the lack of identiﬁcation and treatment of maternal depression in pediatric settings. Inadequate time, insurance limitations, and lack of available mental health services deter providers from screening for and managing maternal depression (Olson et al., 2002).

CONCLUSIONS Women of childbearing age interact with a number of services and systems, but none have assumed responsibility for the identiﬁcation and treatment of maternal depression. Media campaigns have largely failed to inform the public and physicians of the importance of recognizing and treating maternal depression. Most state departments of health do not make an effort to increase awareness of the importance of screening for depression, especially in new mothers. An expansion of the PRAMS program including the depression questions to all 50 states would be an important next step. Sectors capable of screening such as religious organizations, places of employment, childcare centers, and home visiting programs have not been mobilized to implement screening programs and represent important untapped resources. The fragmentation of medical services for young women has also contributed to the lack of identiﬁcation and treatment of maternal depression. Young women use few primary care services, and attention to maternal depression is relatively recent in the OB/GYN literature. In fact, 59% of OB/GYNs do not believe identiﬁcation and treatment is their responsibility, and 66% of OB/GYNs are not conﬁdent in their ability to manage depression. Furthermore, the medical services that have access to mothers with young children focus only on children. Forty-three percent of pediatricians do not believe identiﬁcation and treatment of maternal depression is their responsibility, 55% are not conﬁdent in their ability to diagnose, and 95% are not conﬁdent in their ability to manage this problem. To improve low rates of identiﬁcation and treatment, services interacting with women of childbearing age need to be mobilized. Until these services are sensitized to the need for identiﬁcation and treatment, maternal depression will continue to be underdiagnosed and untreated.

The Failure of Community Settings

25

REFERENCES ACOG Committee Opinion No. 292. (2003). Primary and preventive care: Periodic assessments. Obstetrics and Gynecology, 102(5), 1117–1124. Alvidrez, J., & Azocar, F. (1999). Self-recognition of depression in public care women’s clinic patients. Journal of Women’s Health and Gender-Based Medicine, 8(8), 1063–1071. Bagedahl-Strindlund, M., & Monsen, B. K. (1998). Postnatal depression: A hidden illness. Acta Psychiatrica Scandinavica, 98, 272–275. Bebbington, P., Dunn, G., Jenkins, R., Lewis, G., Brugha, T., Farrell, M., & Meltzer, H. (2003). The inﬂuence of age and sex on the prevalence of depressive conditions: Report from the National Survey of Psychiatric Morbidity. International Review of Psychiatry, 15(1–2), 74–83. Beck, L. F., Morrow, B., Lipscomb, L. E., Johnson, C. H., Gafﬁeld, M. E., Rogers, M., & Gilbert, B. C. (2002). Prevalence of selected maternal behaviors and experiences, pregnancy risk assessment monitoring system (PRAMS), 1999. CDC, 51, 1–26. Behrman, R. E. (1999). Home visiting: Recent program evaluations – analysis and recommendations. The Future of Children, 9(1), 4–26. Behrman, R. E., Gomby, D. S., Larner, M. B., Stevenson, C. S., Lewit, E. M., & Behrman, R. E. (1995). Long-term outcomes of early childhood programs: Analysis and recommendations. The Future of Children, 5(3), 6–24 Review. Blazer, D. G., Kessler, R. C., McGonagle, K. A., & Swartz, M. S. (1994). The prevalence and distribution of major depression in a national community sample: The National Comorbidity survey. American Journal of Psychiatry, 151(7), 979–986. Blehar, M. C., & Oren, D. A. (1997). Gender differences in depression. Medscape Womens Health, 2(2), 3. Bryan, T. L., Georgiopoulos, A. M., Harms, R. W., Huxsahl, J. E., Larson, D. R., & Yawn, B. P. (1999). Incidence of postpartum depression in Olmsted County, Minnesota – A population-based, retrospective study. Journal of Reproductive Medicine, 44, 351–358. Buekens, P., van Heeringen, K., Boutsen, M., Smekens, P., & Mattellaer, P. (1998). Depressive symptoms are often unrecognized in gynaecological practice. European Journal of Obstetrics Gynecology, 81, 43–45. Buist, A., Westley, D., & Hill, C. (1999). Antenatal prevention of postnatal depression. Archives of Women’s Mental Health, 1, 167–173. Chaudron, L. H., Klein, M. H., Remington, P., Palta, M., Allen, C., & Essex, M. J. (2001). Predictors, prodromes and incidence of postpartum depression. Journal of Psychosomatic Obstetrics Gynecology, 22, 103–112. Cheng, T. L., Perrin, E. C., DeWitt, T. G., & O’Connor, K. G. (1996). Use of checklists in pediatric practice. Archives of Pediatrics and Adolescent Medicine, 150(7), 768–769. Cox, J. L., Conner, Y., & Kendell, R. E. (1982). Prospective study of the psychiatric disorders of childbirth. British Journal of Psychiatry, 140, 111–117. Cox, J. L., Murray, D., & Chapman, G. (1993). A controlled study of the onset, duration and prevalence of postnatal depression. British Journal of Psychiatry, 163, 27–31. Cyranowski, J. M., Frank, E., Young, E., & Shear, M. K. (2000). Adolescent onset of the gender difference in lifetime rates of major depression: A theoretical model. Archives of General Psychiatry, 57(1), 21–27. Davies, B. R., Howells, S., & Jenkins, M. (2003). Early detection and treatment of postnatal depression in primary care. Journal of Advanced Nursing, 44(3), 248–255.

26

SARAH MCCUE HORWITZ ET AL.

Druss, B. G., Rosenheck, R. A., & Sledge, W. H. (2000). Health and disability costs of depressive illness in a major U.S. corporation. American Journal of Psychiatry, 157(8), 1274–1278. Duggan, A. K., McFarlane, E. C., Windham, A. M., Rohde, C. A., Salkever, D. S., Fuddy, L., Rosenberg, L. A., Buchbinder, S. B., & Sia, C. C. (1999). Evaluation of Hawaii’s Healthy Start Program. The Future of Children, 9(1), 66–90. Eisenberg, L. (1992). Child mental health in the Americas: A public health approach. Bulletin of the Pan American Health Organization, 26(3), 130–141. Elinson, L., Houck, P., Marcus, S. C., & Pincus, H. A. (2004). Depression and the ability to work. Psychiatric Services, 55, 29–34. Elliott, S. A., Leverton, T. J., Sanjack, M., Turner, H., Cowmeadow, P., Hopkins, J., & Bushnell, D. (2000). Promoting mental health after childbirth: A controlled trial of primary prevention of postnatal depression. British Journal of Clinical Psychology, 39, 223–241. Epperson, C. N. (1999). Postpartum major depression: Detection and treatment. American Family Physician, 59(8), 2259–2260. Federal Interagency Forum on Child and Family Statistics. (2001). http://www.childstats.gov/ Florida State University Center for Prevention & Early Intervention Policy. (2002). Improving Mental Health Services for Perinatal Women. http://www.cpeip.fsu.edu/resourceFiles/ resourceFile_19.pdf French, M. T., Zarkin, G. A., Bray, J. W., & Hartwell, T. D. (1999). Cost of employee assistance programs: Comparison of national estimates from 1993 and 1995. Journal of Behavioral Health Services and Research, 26(1), 95–103. Georgiopoulos, A. M., Bryan, T. L., Woolan, P., & Yawn, B. P. (2001). Routine screening for postpartum depression. Journal of Family Practice, 50(2), 117–122. Georgiopoulos, A. M., Bryan, T. L., Yawn, B. P., Houston, M. S., Rummans, T. A., & Therneau, T. M. (1999). Population-based screening for postpartum depression. Journal of Obstetrics Gynecology, 93(5), 117–122. Good, M. J., Good, B. J., & Cleary, P. D. (1987). Do patient attitudes inﬂuence physician recognition of psychosocial problems in primary care? Journal of Family Practice, 25(1), 53–59. Goodman, S. H., & Gotlib, I. H. (1999). Risk for psychopathology in the children of depressed mothers: A developmental model for understanding mechanisms of transmission. Psychological Review, 106(3), 458–490. Grace, S. L., Evindar, A., & Stewart, D. E. (2003). The effect of postpartum depression on child cognitive development and behavior: A review and critical analysis of the literature. Archives of Women’s Mental Health, 6, 263–274. Green, M. (1994). Diagnosis, management, and implications of maternal depression for children and pediatricians. Current Opinion in Pediatrics, 6, 525–529. Green, M., & Palfrey, J. S. (Eds) (2002). Bright futures: Guidelines for health supervision of infants, children, and adolescents (2nd ed.). Arlington, VA: National Center for Education in Maternal and Child Health. Greenﬁeld, S. F., Reizes, J. M., Muenz, L. R., Kopans, B., Kozloff, R. C., & Jacobs, D. G. (2000). Treatment for depression following the 1996 National Depression Screening Day. American Journal of Psychiatry, 157, 1867–1869. Guttmann, A., Dick, P., & To, T. (2004). Infant hospitalization and maternal depression, poverty and single parenthood: A population-based study. Child Care Health and Development, 30(1), 67–75.

The Failure of Community Settings

27

Hankin, B. L., Abramson, L. Y., Mofﬁtt, T. E., Silva, P. A., McGee, R., & Angell, K. E. (1998). Development of depression from preadolescence to young adulthood: Emerging gender differences in a 10-year longitudinal study. Journal of Abnormal Psychology, 107(1), 128–140. Hayes, B. A., Muller, R., & Bradley, B. S. (2000). Perinatal depression: A randomized controlled trial of an antenatal education intervention for primiparas. BIRTH, 28(1), 28–35. Heneghan, A. M., Silver, E. J., Bauman, L. J., & Stein, R. E. (2000). Do pediatricians recognize mothers with depressive symptoms? Pediatrics, 106(6), 1367–1373. Heneghan, A. M., Mercer, M., & DeLeone, N. L. (2004). Will mothers discuss parenting stress and depressive symptoms with their child’s pediatrician? Pediatrics, 113(3 Pt 1), 460–467. Herrick, H. (2002). Postpartum depression: Who gets help? North Carolina Division of Public Health. ‘‘HHS Awards $2.9 Million To Support 30 Head Start Innovation Projects’’ HHS News 25 March 2004. Hirschfeld, R. M., Keller, M. B., Panico, S., Arons, B. S., Barlow, D., Davidoff, F., Endicott, J., Froom, J., Goldstein, M., Gorman, J. M., Marek, R. G., Maurer, T. A., Meyer, R., Phillips, K., Ross, J., Schwenk, T. L., Sharfstein, S. S., Thase, M. E., & Wyatt, R. J. (1997). The National Depressive and Maniac-Depressive Association Consensus Statement on the undertreatment of depression. Journal of the American Medical Association, 277(4), 333–340. Holden, J. M., Sagovsky, R., & Cox, J. L. (1989). Counseling in a general practice setting: Controlled study of health visitor intervention in treatment of postnatal depression. British Medical Journal, 298, 223–226. Hsiao, M. C., Liu, C. Y., Chen, K. C., & Hsieh, T. T. (2002). Characteristics of women using a mental health clinic in a gynecologic out-patient setting. Psychiatry and Clinical Neurosciences, 56, 459–463. Kahn, R. S., Wise, P. H., Finkelstein, J. A., Bernstein, H. H., Lowe, J. A., & Homer, C. J. (1999). The scope of unmet maternal health needs in pediatric settings. Pediatrics, 103(3), 576–581. Katon, W., von Korff, M., Lin, E., Bush, T., & Ormel, J. (1992). Adequacy and duration of antidepressant treatment in primary care. Medical Care, 30, 20–23. Katon, W., von Korff, M., Lin, E., Simon, G., Walker, E., Bush, T., & Ludman, E. (1997). Collaborative management to achieve depression treatment guidelines. Journal of Clinical Psychiatry, 58(Suppl. 1), 20–23. Kessler, R. C., Barber, C., Birnbaum, H. G., Frank, R. G., Greenberg, P. E., Rose, R. M., Simon, G. E., & Wang, P. (1999). Depression in the workplace: Effects on short-term disability. Health Affairs (Millwood), 18(5), 163–171. Kessler, R. C., McGonagle, K. A., Nelson, C. B., Hughes, M., Swartz, M., & Blazer, D. G. (1994a). Sex and depression in the National comorbidity survey. II: Cohort effects. Journal of Affective Disorders, 30(1), 15–26. Kessler, R. C., McGonagle, K. A., Swartz, M., Blazer, D. G., & Nelson, C. B. (1993). Sex and depression in the National Comorbidity survey. I: Lifetime prevalence, chronicity and recurrence. Affective Disorders, 29(2–3), 85–96. Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., Wittchen, H. U., & Kendler, K. S. (1994b). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry, 51(1), 8–19.

28

SARAH MCCUE HORWITZ ET AL.

Klinkman, M. S. (1997). Competing demands in psychosocial care – Model for the identiﬁcation and treatment of depressive disorders in primary care. General Hospital Psychiatry, 19, 98–111. Klinkman, M. S., Coyne, J. C., Gallo, S., & Schwenk, T. L. (1998). False positives, false negatives, and the validity of the diagnosis of major depression in primary care. Archives of Family Medicine, 7(5), 451–461. Kosmin, B. A., & Mayer, E. (2001). American Religious Identification Survey. New York: The Graduate Center at CUNY: Research Studies. Kumar, R. (1994). Postnatal mental illness: A transcultural perspective. Social Psychiatry and Psychiatric Epidemiology, 29, 250–264. Kumar, R., & Robson, K. M. (1984). A prospective study of emotional disorders in childbearing women. British Journal of Psychiatry, 144, 35–47. Lanzi, R. G., Pascoe, J. M., Keltner, B., & Ramey, S. L. (1999). Correlates of maternal depressive symptoms in a national Head Start program sample. Archives of Pediatrics and Adolescent Medicine, 153, 801–807. LaRocco-Cockburn, A., Melville, J., Bell, M., & Katon, W. (2003). Depression screening attitudes and practices among obstetrician–gynecologists. Obstetrics and Gynecology, 101(5 Pt 1), 892–898. Lipscomb, L. E., Johnson, C. H., Morrow, B., Gilbert, B. C., Ahluwalia, I. B., Beck, L. F., Gafﬁeld, M. E., Rogers, M., & Whitehead, N. (2000). PRAMS 1998 Surveillance Report. Atlanta: Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion, Center for Disease Control and Prevention. Lovejoy, M. C., Graczyk, P. A., O’Hare, E., & Neuman, G. (2000). Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review, 20(5), 561–592. Lumley, J., & Austin, M. P. (2001). What interventions may reduce postpartum depression. Current Opinion in Obstetrics and Gynecology, 13(6), 605–611 Review. Markens, S., Fox, S. A., Taub, B., & Gilbert, M. L. (2002). Role of Black churches in health promotion programs: Lessons from the Los Angeles Mammography Promotion in Churches Program. American Journal of Public Health, 92(5), 805–810. Marlowe, J. F. (2002). Depression’s surprising toll on worker productivity. Employee Benefits Journal, 27(1), 16–21. Maternity Leave and Employment Patterns, 1961–1995. U.S. Department of Commerce Census Bureau. http://www.census.gov Matthey, S., Kavanagh, D. J., Howie, P., Barnett, B., & Charles, M. (2004). Prevention of postnatal distress or depression: An evaluation of an intervention at preparation for parenthood classes. Journal of Affective Disorders, 79, 113–126. McIntosh, J. (1993). Postpartum depression: Women’s help-seeking behaviour and perceptions of cause. Journal of Advanced Nursing, 18(2), 178–184. McLennan, J. D., Kotelchuck, M., & Cho, H. (2001). Prevalence, persistence, and correlates of depressive symptoms in a national sample of mothers of toddlers. Journal of the American Academy Child Adolescent Psychiatry, 40(11), 1316–1323. Miranda, J., Azocar, F., Komaromy, M., & Golding, J. M. (1998). Unmet mental health needs of women in public-sector gynecologic clinics. American Journal of Obstetrics and Gynecology, 178(2), 212–217. Mirola, W. A. (1999). A refuge for some: Gender differences in the relationship between religious involvement and depression. Sociology of Religion, 60(4), 419–437.

The Failure of Community Settings

29

Murray, C. J., & Lopez, A. D. (1996). Evidence-based health policy – Lessons from the Global Burden of Disease Study. Science, 274(5288), 740–743. National Center for Health Statistics. National Health Interview Survey. 2000. Available at: http://www.cdc.gov/nchs/nhis.htm National Depression Screening Day. http://www.mentalhealthscreening.org/depression.htm National Mental Illness Awareness Week. US DHHS: SAMHSA. http://www.mentalhealth. samhsa.gov/highlights/october2003/awareness/ National Public Education Campaign on Clinical Depression. http://www.nmha.org/ NIMH DART (Depression Awareness Recognition and Treatment Program) U.S. Dept. Health and Human Services.{42.1 (Sept. 1995)} http://www.nimh.nih.gov/ NIMH Research Center on Poverty, Risk & Mental Health. (2002). Social Risks and Psychological Health: Strengthening Mental Health Screening in Head Start Programs: The Family Development Project. University of Michigan School of Social Work. O’Hara, M. W., Zekoski, E. M., Phillips, L. H., & Wright, E. J. (1990). Controlled prospective study of postpartum mood disorders: Comparison of childbearing and non-childbearing women. Journal of Abnormal Psychology, 99(1), 3–15. Oklahoma State Department of Health. (1995). Depression after delivery among Oklahoma mothers. PRAMS-Gram, 5(3), 1–4. http://www.health.state.ok.us/program/mchp&e/pramarch Olfson, M., Shea, S., Feder, A., Fuentes, M., Nomura, Y., Gameroff, M., & Weissman, M. M. (2000). Prevalence of anxiety, depression, and substance use disorders in an urban general medicine practice. Archives of Family Medicine, 9, 876–883. Olson, A. L., Kemper, K. J., Kelleher, K. J., Hammond, C. S., Zuckerman, B. S., & Dietrich, A. J. (2002). Primary care pediatricians’ roles and perceived responsibilities in the identiﬁcation and management of maternal depression. Pediatrics, 110(6), 1169–1176. Ormel, J., Koeter, M. W., van den Brink, W., & van de Willige, G. (1991). Recognition, management, and course of anxiety and depression in general practice. Archives of General Psychiatry, 48, 700–706. Oss, M. E., & Clary, J. (1998). The evolving world of employee assistance. Behavioral Health Management, 18(4), 20–24. Parke, S., & Hardy, B. (1997). Postnatal depression: Making better use of health visitors and community psychiatric nurses. Professional Care of Mother and Child, 7(6), 151–152. Pe´rez-Stable, E. J., Miranda, J., Mun˜oz, R. F., & Ying, Y. W. (1990). Depression in medical outpatients: Underrecognition and misdiagnosis. Archives of Internal Medicine, 150, 1083–1088. Pignone, M. P., Gaynes, B. N., Rushton, J. L., Burchell, C. M., Orleans, C. T., Mulrow, C. D., & Lohr, K. N. (2002). Screening for depression in adults: A summary of the evidence for the U.S. Preventive Services Task Force. Annals of Internal Medicine, 136(10), 765–776. Postpartum Depression Resolution (H. J. Res. 163). 1999. Ramey, C. T., Ramey, S. L., Phillips, M., Lanzi, R. G., Brezausek, C., Katholi, C. R., & Snyder, S. (2000). Head Start Children’s Entry into Public School: A report on the National Head Start/Public Early Childhood Transition Demonstration. Civitan International Research, The University of Alabama at Birmingham. Ray, K. L., & Hodnett, E. D. (2001). Caregiver support for postpartum depression. Cochrane Database of Systems Reviews, 3, CD000946. 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. Arch Gen Psychiatry, 50(2), 85–94.

30

SARAH MCCUE HORWITZ ET AL.

Reid, A. J., Biringer, A., Carroll, J. D., Midmer, D., Wilson, L. M., Chalmers, B., & Stewart, D. E. (1998). Using the ALPHA form in practice to assess antenatal psychosocial health. Canadian Medical Association Journal, 159(6), 677–684. Reynes, R. (1998). Programs that aid troubled workers. Nation’s Business, 86(6), 73–74. Schmidt, L. A., Greenberg, B. D., Holtzman, G. B., & Schulkin, J. (1997). Treatment of depression by obstetrician-gynecologists: A survey study. Obstet Gynecol, 90(2), 296–300. Seidman, D. (1998). Postpartum psychiatric illness: The role of the pediatrician. Pediatr Rev, 19(4), 128–131. Spitzer, R. L., Williams, J. B., Kroenke, K., Hornyak, R., & McMurry, J. (2000). Validity and utility of the PRIME-MD patient health questionnaire in assessment of 3000 obstetric– gynecologic patients: The PRIME-MD Patient Health Questionnaire Obstetrics– Gynecology Study. American Journal of Obstetrics and Gynecology, 183(3), 759–769. Sprang, G. (1992). Utilizing a brief EAP-based intervention as an agent for change in the treatment of depression. Employee Assistance Quarterly, 8(2), 57–65. Stevens, J. C., & Diehl, S. J. (2003). OB/GYN residents as primary care providers: Implementing a new curriculum for diagnosing and treating depression and anxiety. Primary Care Update OB/GYNS, 10, 297–299. Substance Abuse and Mental Health Services Administration. (2000). Mental health: A report of the Surgeon General. Washington, DC: Department of Health and Human Services. Sundstro¨m, I. M., Bixo, M., Bjo¨rn, I., & Astrom, M. (2001). Prevalence of psychiatric disorders in gynecologic outpatients. American Journal of Obstetrics and Gynecology, 184(2), 8–13. The Center for Maternal and Child Health, Maryland Department of Health and Mental Hygiene, and Health Care Answers. The Health of Maryland Women 2002. Baltimore, Maryland, 2002. http://www.fha.state.md.us/mch/pdf/WomensHealth-Publication.pdf U.S. Dept. of Commerce Census Bureau. http://www.census.gov U.S. Department of Health and Human Services, Administration for Children and Families, Head Start Bureau. (2002). Services to pregnant women participating in early head start Bureau. Issued March 27, 2002. http://www.headstartinfo.org/publications/im02/ im02_04a.htm U.S. Department of Health and Human Services, Agency for Health Care Policy and Research. (1993). Depression in primary care: Treatment of major depression (Vol. 2). Rockville, MD: The Agency. U.S. Preventive Services Task Force. (2002). Screening for Depression: Recommendations and Rationale. May 2002. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/clinic/3rduspstf/depressrr.htm U.S. Public Health Service. (2000). Report of the Surgeon General’s Conference on Children’s Mental Health: A national action agenda. Washington, DC: Department of Health and Human Services. Weissman, M. M., & Klerman, G. L. (1977). Sex differences and the epidemiology of depression. Archives of General Psychiatry, 34(1), 98–111. Weitzman, M., & Auinger, P. (2002). Personal Communication. Wells, K. B., Sherbourne, C., Schoenbaum, M., Duan, N., Meredith, L., Unutzer, J., Miranda, J., Carney, M. F., & Rubenstein, L. V. (2000). Impact of disseminating quality improvement programs for depression in managed primary care. Journal of the American Medical Association, 283(2), 212–220.

The Failure of Community Settings

31

Wells, K. B., Shoenbaum, M., Unu¨tzer, J., Lagomasino, I. T., & Rubenstein, L. V. (1999). Quality of care for primary care patients with depression in managed care. Archives of Family Medicine, 8, 529–536. Wells, K. B., Sturm, R., Sherbourne, C. D., & Meredith, L. S. (1996). Caring for depression. Cambridge, MA: Harvard University Press. Wessel, R., & The Ellsworth Associates Research Team. (2000). Maternal depression: A summary of current literature and annotated bibliography. Whitton, A., & Appleby, L. (1996). Maternal thinking and the treatment of postnatal depression. International Review of Psychiatry, 8(1), 73–78. Williams, J. W., Jr., Rost, K., Dietrich, A. J., Ciotti, M. C., Zyzanski, S. J., & Cornell, J. (1999). Primary care physicians’ approach to depressive disorders. Archives of Family Medicine, 8(1), 58–67. Wisner, K. L., Parry, B. L., & Piontek, C. M. (2002). Postpartum depression. New England Journal of Medicine, 347(3), 194–199. Wisner, K. L., Zarin, D. A., Holmboe, E. S., Appelbaum, P. S., Gelenberg, A. J., Leonard, H. L., & Frank, E. (2000). Risk-beneﬁt decision making for treatment of depression during pregnancy. American Journal of Psychiatry, 157(12), 1933–1940. Yonkers, K. A., & Chantilis, S. J. (1995). Recognition of depression in obstetric/gynecology practices. American Journal of Obstetrics and Gynecology, 173(2), 632–638. Yonkers, K. A., Ramin, S. M., Rush, A. J., Navarrete, C. A., Carmody, T., March, D., Heartwell, S. F., & Leveno, K. J. (2001). Onset and persistence of postpartum depression in an inner-city maternal health clinic system. American Journal of Psychiatry, 158(11), 1856–1863. Zlotnick, C., Johnson, S. L., Miller, I. W., Pearlstein, T., & Howard, M. (2001). Postpartum depression in women receiving public assistance: Pilot study of an interpersonal therapyoriented group intervention. American Journal of Psychiatry, 158, 638–640. Zuckerman, B. S., & Beardslee, W. R. (1987). Maternal depression: A concern for pediatricians. Pediatrics, 79(1), 110–117.

SCHOOL-BASED MENTAL HEALTH SERVICES FOR CHILDREN AND ADOLESCENTS Melissa Pearrow and Peter Whelley SCHOOL-BASED MENTAL HEALTH SERVICES FOR CHILDREN AND ADOLESCENTS Public schools possess a unique constellation of opportunities and challenges for mental health service provision. Schools, as settings within a larger ecological context, can be a community institution that supports a child as s/he develops assets for resilient development while providing opportunities for a range of life choices. School is the setting where children can learn and practice peer relations and social norms, and it can be a refuge where children who have many environmental risks can ﬁnd structure and effective methods of success (Doll, 1999). When Willie Horton, the infamous bank robber, was asked why he robbed banks, he responded, ‘‘Because that’s where the money is.’’ At a most basic level, schools are where the children are. Every day more than 52 million students attend over 1,14,000 schools in the United States, and including the 6 million adult staff, this amounts to almost one-ﬁfth of the population passing through the Nation’s schools on any given weekday (New Freedom Commission on Mental Health, 2003). The societal mandate placed upon institutions of public education is to provide a foundation of academic skills, though many psychological, social, Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 33–51 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14003-9

33

34

MELISSA PEARROW AND PETER WHELLEY

and emotional barriers can impact a child’s engagement in learning, which can complicate opportunities for future success. There is a wide range of social and psychological barriers that can impair a child’s ability to perform in the structured school environment, including impulse control deﬁcits, poverty, homelessness, racism, and familial mental illness. As such, most schools provide an array of services at various age levels that are relevant to mental health research. Multiple studies in peer-reviewed journals and professional publications demonstrate the effectiveness of mental health services in schools. For example, over 100 evidence-based mental health programs implemented within the school setting are reported in the Exemplary Mental Health Programs: School Psychologists as Mental Health Service Providers (Nastasi, Plymert, Varjas, & Moore, 2002). This chapter will introduce some of the special issues that impact the work of mental health providers and social scientists in the public education environment. It begins with the historical developments and legal parameters of the American public education system, speciﬁcally as they impact mental health services in the schools. Children and adolescents identiﬁed as having the greatest mental health needs will also be addressed, speciﬁcally as related to the supports and mandates of the educational systems. The recent trend in mental health delivery through school-based mental health centers will also be examined. Given the contextual focus of this chapter, there will also be a discussion of how schools are an ideal setting for prevention programming, based on a public health model, and an examination of future directions for service provision in this unique setting.

HISTORY OF PUBLIC EDUCATION AND ITS MANDATES Public education systems are a ‘‘deeply political enterprise’’ and have long reﬂected trends in the broader society in which they operate (Hochschild & Scovronick, 2003, p. xi). An outcome of women’s suffrage in the late 1800s and early 1900s was expanded educational opportunities. One of the landmark civil rights rulings, Brown versus the Board of Education in 1954, authorized the education and integration of all students regardless of race. Reﬂecting the activism and community inclusion of the 1960s and 1970s, the Education of All Handicapped Children Act, P.L. 94-142, later reauthorized as the Individuals with Disabilities Education Act (IDEA), was passed in 1975 to highlight that all children, even those with disabilities, had the right to be educated. These historical incidents provide the background for the

School-Based Mental Health Services

35

current public education system that continues to address the educational, social, and psychological needs of children and adolescents. The initial federal law of 1975 provided the framework of special education and determined the overarching regulations implemented at the state level by the Departments of Education. It deﬁned speciﬁc disabilities that qualify a student for special education services, and required that students with disabilities be placed in the least restrictive environment (LRE) in which educational needs could be satisfactorily served. The intent of this education law was to remove the stigma attached to students with disabilities, improve social relationships, provide non-disabled learning models, and prepare for living in the real world outside of school (Pulliam & Van Patten, 1995). However, like many legal mandates, the law is dynamic and requires revisions over time. The most recent reauthorization, the Individuals with Disabilities Education Improvement Act (IDEIA) effective from July 2005, again alters the criteria that determine eligibility for special education services that in turn alter the practices of regular and special education personnel. Furthermore, other federal legislation, such as No Child Left Behind (NCLB), critically examines the academic progress of all students and again alters the day-to-day practices of school personnel. Since the inception of this federal law in 1975, signiﬁcant progress has been made toward meeting the national goals for implementing effective educational programs and services (U.S. Department of Education, n.d.a). Compared to 1970, when only one in ﬁve children with disabilities were educated in public schools, striking progress has been made in the services and programs where the majority of children with disabilities are being educated in their neighborhood schools. The federal law guarantees a free and appropriate education (FAPE) in the LRE for children between the ages of 3 and 21. In 1986, the Amendment P.L. 99-457 mandated that public education systems also provide services to qualifying children between the ages 0 and 3 (U.S. Department of Education, n.d.a). The federal law, currently referred to as IDEA, affords deﬁnitions that determine eligibility for special education services. It indicates that the term ‘‘child with a disability’’ means a child (i) with mental retardation, hearing impairments (including deafness), speech or language impairments, visual impairments (including blindness), serious emotional disturbance (hereinafter referred to as emotional disturbance (ED)), orthopedic impairments, autism, traumatic brain injury, other health impairments, or speciﬁc learning disabilities; and (ii) who, by reason thereof, needs special education and related services (Council for Exceptional Children, n.d.). Each of these eligibility categories is deﬁned in IDEA with speciﬁc regulations outlined by

36

MELISSA PEARROW AND PETER WHELLEY

each state; thus, a degree of variability exists between states in their eligibility criteria and service provision. There are challenges in the process of determining eligibility for special education services. For example, the Surgeon General’s Report on Mental Health (1999) indicated that approximately 20% of children and adolescents demonstrate symptoms qualifying them for diagnoses under the Diagnostic and Statistical Manual – Fourth Edition (DSM-IV: American Psychiatric Association, 1999); however, this percentage is inconsistent with the number of students provided special education services within the public schools. Based on 2003 population estimates of children between the ages of 6 and 17 served under IDEA, the average number of students receiving special education services is approximately 11.46%, with state averages ranging from 16.51% in Rhode Island to 9.01% in Colorado (U.S. Department of Education, Ofﬁce of Special Education Programs, 2004). These percentages fall short of the 20% of children and adolescents who demonstrate enough symptoms to receive psychiatric diagnoses. Although the reasons for these differences are complex, one basic explanation is the difference in language between the mental health and the educational communities. For example, the term ‘‘emotional disturbance’’ is not a formal psychiatric diagnostic category but is a term used in several federal statutes. Thus, this term does not signify a particular diagnosis, but instead becomes a legal term that prompts a series of services for children and youth in the educational community. Special education law clearly stipulates that having a disability alone does not warrant eligibility for special education services, and requires that the disability must impact the child’s ability to make effective academic progress. The differences in the mission and professional language of educational and mental health systems can create confusion and conﬂict, especially since psychiatric diagnoses do not always translate into special education services. Examples of diagnoses that would not necessarily lead to special education services can include disruptive behavior disorders, such as attention deﬁcit disorder (with or without hyperactivity), oppositional deﬁant disorder, or conduct disorder. Such psychiatric diagnoses may not impact academic development and the child may continue to make ‘‘adequate’’ academic progress, thus not making them eligible for special education services. While the current laws restrict access to special education services to those who ‘‘qualify,’’ other avenues of support services are available and accessible to children with disabilities. Civil rights laws (e.g., Section 504) require ‘‘equal access’’ for individuals with disabilities, and this has resulted in accommodation plans that outline what support services are necessary in

School-Based Mental Health Services

37

order for the child to access the curriculum (Pulliam & Van Patten, 1995). For example, a child could have an attention deﬁcit disorder and be making effective academic progress but still requires accommodations to have equal access to the curriculum. These accommodations vary based on the individual needs of the children or adolescent but are available to any child with a disability. Many of the previously mentioned categorical disabilities that qualify a child for special education services would indicate that mental health support services seem necessary to ensure academic development, such as ED or autism, while others would not necessarily require mental health support services, such as hearing/visual or orthopedic impairments (though there are clear arguments for the beneﬁts of support services to assist in life adjustment). Mental health services – be they individual or group counseling or behavioral interventions – provided by school support personnel can be a part of a child’s individualized educational program (IEP) if the team determines that they are necessary to support the academic progress of the child. As such, nearly half of all schools offer social work or psychological services to students with disabilities, whether provided by school personnel or outside agency providers, and nearly 70% of schools have collaborative agreements with community mental health agencies (Bradley, Henderson, & Monfore, 2004). There are some students who clearly bridge the mental health and education worlds due to their disability with chronic mental illness. For these students, the category of ED in IDEA includes provisions for special education services. The needs of this population generally require comprehensive service provision from mental health professionals both inside and outside of the school system. As such, the following is a closer examination of the educational eligibility criteria for ED, issues concerning collaboration between schools and mental health agencies, and evidence-based practices with this unique population.

STUDENTS IDENTIFIED AS HAVING AN EMOTIONAL DISTURBANCE (ED) The provision of special education services under the primary disability of ED is rare. In 2003, a total of 483,805 students between the ages of 6 and 21 were served under IDEA with the primary disability of ED (U.S. Department of Education, Ofﬁce of Special Education Programs, 2004), making up approximately 8% of all students identiﬁed as having a disability. Special

38

MELISSA PEARROW AND PETER WHELLEY

education law stipulates the criteria that must be met to qualify for services under the category of ED. The criteria include the following: (i) The term means a condition exhibiting one or more of the following characteristics over a long period of time and to a marked degree that adversely affects a child’s educational performance: (A) An inability to learn that cannot be explained by intellectual, sensory, or health factors. (B) An inability to build or maintain satisfactory interpersonal relationships with peers and teachers. (C) Inappropriate types of behavior or feelings under normal circumstances. (D) A general pervasive mood of unhappiness or depression. (E) A tendency to develop physical symptoms or fears associated with personal or school problems. (ii) The term includes schizophrenia. The term does not apply to children who are socially maladjusted, unless it is determined that they have an emotional disturbance (Code of Federal Regulations, Title 34, Section 300.7 (c)(4)(i)). Case law and court decisions have deﬁned salient features of this deﬁnition; for example, ‘‘over a long period of time’’ has generally been deﬁned as a period of six months or more. Debates regarding the deﬁnition of ‘‘socially maladjusted’’ have indicated that disruptive behavior disorders, such as conduct disorder or oppositional deﬁant disorder (American Psychiatric Association, 1999), do not compulsorily qualify under the category of ED. Thus, according to this deﬁnition, students who display some of the more disruptive and maladaptive behavior patterns may not qualify for special education supports and protections. Some legal counsel has suggested that a licensed member of the mental health community must provide, in writing, a diagnosis that meets one of the aforementioned criteria. Yet, cases that have been mediated through local departments of education have indicated that the eligibility criteria can be met – whether or not agencies providers are involved with the child. This is one of the complexities that school systems wrestle with as they meet the education needs of this unique population. There are distinctive environmental and educational risk factors with children and adolescents classiﬁed with ED, according to national studies by Wagner, Kutash, Duchnowski, Epstein, and Sumi (2005) and Bradley et al. (2004). Seventy-ﬁve to 80% of the students with ED are male and a disproportional number, roughly 25%, are African-American (Bradley et al., 2004).

School-Based Mental Health Services

39

The demographic variables of students with ED have not changed from 1987 to 2001, with the exception of a signiﬁcant increase in students who had English as a second language (Bradley et al., 2004). Approximately one-third live in poverty and in households headed by a single parent, and one-ﬁfth live in households in which the head of house is unemployed and not a highschool graduate (Wagner et al., 2005). Almost half (45%) live in a household with another person who has a disability (Wagner et al., 2005). On average, they are provided with special education services more than a year later than other students with disabilities (average age 7.8 for children classiﬁed with ED and 6.7 for children with other disabilities) (Bradley et al., 2004). They are more likely to have changed schools often, primarily because they have been reassigned by their school district, rather than because of grade-level progression or, in the case of elementary/middle school children with ED, because their family moved (Wagner et al., 2005). Parents of students identiﬁed as ED are signiﬁcantly more likely to express dissatisfaction with their children’s schools, teachers, and special education services (Wagner et al., 2005). Approximately half of the students identiﬁed with ED are prescribed medication, primarily stimulants and anti-depressants (Bradley et al., 2004), which indicates that service providers other than school personnel are involved with many of these children and adolescents. Approximately 70% of parents of children with ED indicated that the child or family received psychological treatment, mental health, or counseling services, and nearly 44% of these received these mental health services by or through the school (Bradley et al., 2004). The educational, behavioral, and social outcomes of students with ED continue to be the worst of any disability group (Bradley et al., 2004). They have the highest rate of school failure, despite being an educational priority of the Federal Department of Education since the mid-1960s. Students with ED have signiﬁcant deﬁcits in academic achievement (Reid, Gonzalez, Nordness, Trout, & Epstein, 2004) and 50% of them drop out of high school, compared to 30% of all students with disabilities (New Freedom Commission on Mental Health, 2003). Furthermore, it has been estimated that 70% of these students will be arrested within three years of leaving school, thus continuing a pervasive pattern of failure that becomes difﬁcult to correct (U.S. Department of Health and Human Services, 1999). Concerns regarding the ‘‘pipeline’’ from school to prison, speciﬁcally with this population, have recently been highlighted by researchers with more information available from Ward and Losen (2003). The relationship between environmental risk factors and the development of emotional disabilities in young children is well established (Conroy &

40

MELISSA PEARROW AND PETER WHELLEY

Brown, 2004) and these risk factors highlight the need to address emotional development from a systemic perspective. Students with ED generally require a high level of community support services; however, the school system cannot mandate that a child or their family access community or mental health services despite the needs observed in the school setting. In turn, children can be identiﬁed by community agencies as having a serious ED but not receive special education services from the school system. For example, in a study of 182 children and adolescents identiﬁed as having an ED by a mental health agency, only 70% were identiﬁed as having special needs by the educational system (Taub & Pearrow, 2005b). Children identiﬁed as ED and their families present with complex issues that can result in their being identiﬁed as ‘‘non-compliant’’ when left to traditional outpatient mental health service provision (weekly ofﬁce appointments). Fragmented and inadequate service delivery that was historically over-reliant on institutional care has shifted to a ‘‘systems of care’’ model that is designed to provide culturally competent, coordinated services that are community-based and include family participation (U.S. Department of Health and Human Services, 1999). In 1992, the Comprehensive Community Mental Health Services for Children and their Families Program was created (P.L. 102-321), and was based on the ‘‘systems of care’’ model. This model is dedicated to coordinating service providers from multiple systems to develop a cohesive plan to support the child with mental health impairments stay in the home and community, and relies on the family to determine who their ‘natural supports’ are in order to rely less on professionals, as in the typical clinical model (Worthington, Hernandez, Friedman, & Uzzell, 2001). Further information regarding this model is available from Kendziora, Bruns, Osher, Pacchiano, and Mejia (2001) and in other chapters of this book. Success with this population requires family engagement in the delivery of services as a key component to effective implementation (Burns, Hoagwood, & Mrazek, 1999). While many features of a successful relationship between families and service providers are emerging – such as ease of access, responsiveness, trust and acceptance, and continuity of care – there is limited information on the role of schools. Wraparound services, from the ‘‘systems of care’’ concept, are primarily initiated through mental health or child welfare systems, yet can result in improved outcomes for students with ED (Eber, Sugai, Smith, & Scott, 2002). Taub and Pearrow (2005b) report a decrease in maladaptive behaviors though no improvements were indicated in the development of positive behaviors, based on parent ratings, in the school environment. Limited information regarding school functioning can become available due to reluctance and dissatisfaction of family members.

School-Based Mental Health Services

41

This reluctance, combined with privacy protections laws such as Family Educational Rights and Privacy Act (FERPA) and Health Insurance Portability and Accountability Act of 1996 (HIPAA), often restricts access to information that enables further research in this area. When a student is identiﬁed as having an ED, signiﬁcant controversy can arise – one that can markedly disrupt the child’s or adolescent’s future. Being identiﬁed as ED can result in long-term ramiﬁcations such as being denied access to military enrolment, and at times denied security clearances (Olympia, Farley, Christiansen, Pettersen, Jensen, & Clark, 2004). In the shadow of the Patriot Act, American society is in an era where security clearances and background checks are more prevalent. As such, the children who are struggling with ED or mental illness may be at a distinct disadvantage, if not discriminated against, as they seek career options. Clinicians in mental health settings may also experience unique challenges when working with school personnel when a student is identiﬁed as having an ED. Students with ED require signiﬁcant amounts of expertise, time, and resources and can be a destabilizing inﬂuence in a school (Eber et al., 2002). It has been these authors’ experience that generally, many avenues and interventions have been attempted by multiple staff members before a student is identiﬁed as ED. Teachers and support staff may spend signiﬁcant effort in supporting the child in the regular education setting, and may experience a period of mourning and self-doubt when it is determined that the needs of the child exceed the services available to the regular education environment. After this point is reached within the system, many staff members – both teachers and administrators – are reluctant to return these students to regular education settings. This ‘‘black hole’’ scenario, though not openly acknowledged and discussed, can become a part of students with ED educational experiences.

SCHOOL-BASED HEALTH CENTERS: A RECENT TREND As previously mentioned, there is a relatively low incidence of identifying students as ED in the public education system. Nonetheless, the mental health needs of the general school population are acknowledged by community agencies, school administrators, and local politicians. The recommendation to improve and expand school mental health programs was realized in the President’s New Freedom Commission on Mental Health (2003), which states ‘‘mental health is essential to learning as well as to social and emotional development. Because of this important interplay between

42

MELISSA PEARROW AND PETER WHELLEY

emotional health and school success, schools must be partners in the mental health care of our children’’ (p. 58). This report also indicates that school mental health programs can do more than ensure academic achievement: they can also address the health and behavioral concerns of students and reduce unnecessary pain and suffering. As such, large federal support has become available to assist the provision and expansion of school-based health centers to include medical, dental, and mental health services (Adelman & Taylor, 1999; Weist & Evans, 2005). Two national centers of School Mental-Health Assistance are in the University of Maryland School of Medicine and the University of California, Los Angeles. These projects have provided a wealth of information to school personnel regarding the need for these services as well as strategies for implementing them within the school setting. They have identiﬁed several advantages of providing mental health services in schools, including improved educational outcomes by decreasing absences, decreasing discipline referrals, and improving test scores (New Freedom Commission on Mental Health, 2003). Additional advantages of the school setting include that they are a trusted and familiar environment where transportation barriers are reduced or eliminated, in addition to being a natural environment where behavioral difﬁculties impact functioning. They are also a site where early intervention, well-care visits, and screenings can occur without major disruptions to the family. The services provided in a school-based setting can access children and adolescents who may have difﬁculty accessing community resources for various reasons. For example, a study in the Journal of School Health (Blake, Ledsky, Goodenow, & O’Donnell, 2001) indicated that children of immigrant families are more likely to access services when they are provided in school-based health centers. The authors also concluded that ‘‘schools provide a critical gateway and opportunity for reaching both immigrant students and their families with coordinated school health programs’’ (p. 112). These centers become imperative as the Surgeon General’s Report on Mental Health has indicated that minorities have less access to, and availability of, mental health services, often receive a poorer quality of mental health care, and are underrepresented in mental health research (U.S. Department of Health and Human Services, 2001a). School-based health centers have the opportunity to provide secondary prevention services to children or adolescents identiﬁed as displaying ‘‘atrisk’’ behaviors, and allow for earlier involvement of community-based agencies. Examples of groups that have been conducted at a school-based health center that might not fall within the domains of an educational

School-Based Mental Health Services

43

system include substance abuse cessation, smoking prevention, and sex education. Connections to community resources can be provided to the students and families through the school-based centers, which, in theory, allows for sustainability beyond the school hours and school years. Noteworthy concerns with school-based health centers are present, primarily in that they retain the medical model of intervention. Treatment tends to focus on the individual child or adolescent, which can be problematic as this young, dependent population is more vulnerable to its familial environment than adults. While parental/guardian consent is required for treatment, it does not ensure the engagement of the children’s caretakers in the treatment process, as they are no longer arranging and transporting them to appointments. This reduced contact with mental health providers may provide an easy access point to children and adolescents but at the expense of the involvement of caretakers. Anecdotally, a cadre of school administrators has shared their policy of restricting the involvement of external agency mental health staff. These concerns stem from situations where the students, who have great needs and are emotionally vulnerable, engage in treatment in the school setting and then have difﬁculty returning to the demands of school. Another concern heard from school administrators includes situations where treatment providers come to the school to see the child but do not provide consultation to the education staff. While the HIPAA clearly restricts the disclosure of conﬁdential information, it restricts the ability of school professionals who work daily with the child to have information that can assist them while supporting the struggling child or adolescent. Furthermore, mental health services provided by other community-based agencies are usually governed by the limitations and mandates of the insurance companies that compensate the professionals for the services. A discontinuation or termination of services can result in the child being unwilling to re-engage when these limitations are overcome. Yet, these limitations do not absolve the school system of providing for the needs of the students.

SCHOOLS AS SITES FOR MENTAL HEALTH PREVENTION AND IINTERVENTION The mental health services reviewed to this point sustain an ontogenic focus where there is an identiﬁed individual whose pathology needs treatment. The following section will take a paradigmatic shift away from a clinical orientation, where the pathology is perceived to be entirely within the

44

MELISSA PEARROW AND PETER WHELLEY

individual, to a public health orientation, where needs are identiﬁed as population based, and services take on a prevention approach. In discussing schools as sites for prevention, this section will begin with a deﬁnition of prevention, and explain the three-tiered public health model as applied to the school setting, and current research that highlights the need for educational policies and collaborative and integrative programs that address behavior through the school setting (Greenberg et al., 2003). Goldston (1985), former coordinator of prevention programs at the National Institute of Mental Health, deﬁned prevention programs as ‘‘actions taken prior to the onset of disease to intercept its causation, or to modify its course before pathology is involved’’ (p. 453). Bloom (1996) expands on the deﬁnition of primary prevention by including an ecological perspective, in which the programs occur in natural physical and sociocultural settings. By deﬁnition, prevention programs maintain a focus that is educational in nature rather than clinical (Samaraweera & Hurwitz, 2000). Unfortunately, many of the mental health services provided in schools are usually restricted to those legally mandated for students diagnosed with special needs (Adelman & Taylor, 1999; Weist & Evans, 2005). Children exposed to environmental risk factors (e.g., poverty, domestic violence, substance abuse, and abuse and neglect) are at a greater risk for developing emotional and behavioral dysfunction (Conroy & Brown, 2004). Furthermore, educational policies (e.g., IDEA) that require a student to demonstrate signiﬁcant academic failures and delays before services are provided lead to a reactive stance by service providers. Effective prevention programs, such as those targeting tobacco, alcohol and drug abuse, and school drop-out prevention, require comprehensive instruction that begins early in life and stress interventions at the elementary school level, with a focus on both academic and behavioral interventions (Rush & Vitale, 1994). Patterson, DeBaryshe, and Ramsey (1989), in their examination of the developmental progression of anti-social behavior, indicate that prevention should focus on the elementary level and include parent training, child social-skills training, and academic remediation. Petersen, Pietrzak, and Speaker (1998) highlighted the need to target preschool and elementary school-aged children because interventions that target middle schools and high schools have met with limited success. A meta-analysis of prevention programs, conducted by Nelson, Westhues, and MacLeod (2003), reported that preschool prevention programs do have positive effects in the short and the long terms, with the largest outcomes with programs that served predominantly African-American children, who as a population are disproportionately identiﬁed as ED.

School-Based Mental Health Services

45

In Greenberg, Domitrovich, and Bumbarger’s (2001) review of primary prevention programs, the more enduring beneﬁts occur with multi-year programs that are directed at risk and protective factors and link community systems. Successful interventions target developmental levels and include strategies to target the individual and the environment with a focus on school ecology to support the child’s positive changes in both the school and the home environments (Greenberg et al., 2001; Taub & Pearrow, 2005a). Greenberg et al. (2001) offered the following deﬁnition of the three-tiered approach to prevention strategies: ‘‘Universal preventive interventions target the general public or a whole population group that has not been identiﬁed on the basis of individual risk. Exemplars include prenatal care, childhood immunization, and school-based competence enhancement programs. Because universal programs are positive, proactive, and provided independent of risk status, their potential for stigmatizing participants is minimized and they may be more readily accepted and adopted. Selective interventions target individuals or a subgroups (based on biological or social risk factors) whose risk of developing mental disorders is signiﬁcantly higher than average. Examples of selective intervention programs include: home visitation and infant day care for low-birth weight children, preschool programs for all children from poor neighborhoods, and support groups for children who have suffered losses/traumas. Indicated preventive interventions target individuals who are identiﬁed as having prodromal signs or symptoms or biological markers related to mental disorders, but who do not yet meet diagnostic criteria. Providing social skills or parent–child interaction training for children who have early behavioral problems are examples of indicated interventions.’’ (p. 7)

A public health orientation results in community-based strategies that offer practical and goal-oriented interventions for promoting and maintaining health (U.S. Department of Health and Human Services, 2001b). The public health approach to prevention focuses on ‘‘how to change population-level behaviors, environmental factors, or processes to reduce incidence rates of disorders and to increase healthy outcomes in a population’’ (Winslow, Sandler, & Wolchik, 2005, p. 339). This approach has been applied, through the Ofﬁce of the Surgeon General, to address problem behaviors such as youth violence, as it ‘‘invites an approach that focuses more on prevention than on rehabilitation’’ (U.S. Department of Health and Human Services, 2001b, p. viii). According to the Surgeon General’s Report, in using a public health approach to identify problems and develop solutions for entire population groups, the following steps are necessary: Deﬁne the problem, using surveillance processes designed to gather data that establish the nature of the problem and the trends in its incidence and prevalence.

46

MELISSA PEARROW AND PETER WHELLEY

Identify potential causes, through epidemiological analyses that identify risk and protective factors associated with the problem. Design, develop, and evaluate the effectiveness and generalizability of interventions. Disseminate successful models as part of a coordinated effort to educate and reach out to the public (U.S. Department of Health and Human Services, 2001b, p. ix). The focus of interventions becomes the environment rather than the individual. School effectiveness research shows that schools have major effects on children’s development (Johnson, Schwartz, Livingston, & Slate, 2000), thus making them the most logical sites for prevention and early intervention. The educational system offers ‘‘the most efﬁcient and systematic means available to promote the psychological, social, and physical health of school-age children’’ (Weissberg, Caplan, & Harwood, 1991, p. 833). The Consortium on the School-Based Promotion of Social Competence (1996) asserts that schools are the major setting in which activities should take place to promote students’ competence and healthy behavior patterns. Weissberg, Caplan, and Sivo (1989) also advocate for the promotion of social competence within this naturalistic setting – in classrooms and on the playground – where the skills can be developed and generalized and become more effective than efforts utilized in traditional person-centered interventions or through other more external community organizations. Schools also become the logical site for prevention as the largest proportion of state and federal spendings on children in this government system (U.S. Department of Education, n.d.c). For example, the total taxpayer investment in K-12 education in the United States for the 2003–2004 school year exceeded US$ 501.3 billion, primarily funded at state and local levels, and exceeded the national defense budget (U.S. Department of Education, n.d.c). This public health orientation is evident in the reauthorization of the IDEA in 1997 as it includes speciﬁc provisions to address the issues of Positive Behaviors in School (PBIS) within the school environment. PBIS is a systems approach for establishing a continuum of proactive, positive discipline procedures for all students in all types of school settings (Sugai, Sprague, Horner, & Walker, 2000). The U.S. Department of Education (n.d.b) has deﬁned positive behavior supports as ‘‘an application of a behaviorally-based systems approach to enhance the capacity of schools, families, and communities to design effective environments that improve the link between research-validated practices and the environments in which teaching and learning occurs. Attention is focused on creating and sustaining

School-Based Mental Health Services

47

primary (school-wide), secondary (classroom), and tertiary (individual) systems of support that improve lifestyle results (personal, health, social, family, work, recreation) for all children and youth by making problem behavior less effective, efﬁcient, and relevant, and desired behavior more functional.’’ This focus on supports in the general education environment provides structure for the prevention of maladaptive behavior patterns. While all of these arguments suggest the beneﬁts of engaging in activities that promote positive mental health in a preventive capacity, this issue is not without controversy. School attendance is compulsory; however, parents typically do not have a choice in which school their child attends and the curriculum and materials presented. School-wide initiatives that evolve into value lessons can become problematic at best and discriminatory at worst with controversy revolving around whose values will be taught in the schools. Bradsher (1996) indicated that ‘‘value, or character, education has proliferated partly in response to public concerns about the behavior of young people and partly as an alternative to efforts by Christian groups to put prayer and religious values back into schools.’’ Court cases and community disputes challenging these issues can be readily seen in local newspapers. The application of a public health perspective to mental health needs in the school setting offers a reframe of the prevention services; however, with this reframe, caution and controversy appear essential. For example, there are medical issues that, from a public health initiative, restrict access to school, including a lack of vaccinations or having a communicable disease (e.g., lice and impetigo); yet, research also highlights the contagion effective of maladaptive behaviors such as self-mutilation (Hawton, Rodham, Evans, & Weatherall, 2002). A public health approach could provide argument that students who engage in these behaviors should be ‘‘quarantined’’ or restricted from the regular education setting. Despite these points, the expansion of mental health and preventive services in the school setting through a public health model is recommended as research supports the beneﬁts of teaching social and emotional learning and organizing schools in a manner that enhances the development of all children (Elias, Zins, Graczyk, & Weissberg, 2003).

SUMMARY Provision of mental health services at the ontogenic level based on the medical model is needed; however, application of the most state-of-the-art therapeutic interventions is no replacement for preventing problem

48

MELISSA PEARROW AND PETER WHELLEY

behaviors at their inception. From an ecological and public health stance, schools can play a vital role in prevention and early intervention of problem behaviors, especially since some of the most serious health and social problems confronting American society today are caused, in large part, by the behavior patterns established during youth (Kolbe, Collins, & Cortese, 1997). However, this requires a paradigmatic shift of the expectations and mandates on public education systems on behalf of educators, political leaders, and the larger community. A return to the promises from the deinstitutionalization era through the community mental health movement (which utilizes the systems and structures already in place to support adaptive development) is supported with a public health paradigm. The President’s New Freedom Commission on Mental Health (2003) highlights the interplay between emotional health and school success and recommends that schools partner in the mental health care of the nation’s children. While schools are perfect institutions for primary prevention measures, they are also challenged to Leave No Child Behind and to conform to other initiatives du jour. There is more than enough work and fewer and fewer resources available to meet these challenges; for example, within the ﬁeld of school psychology, there is a critical shortage of trained personnel in many geographic regions (Davis, McIntosh, Phelps, & Kehle, 2004). Toward this end, the challenge becomes to effectively collaborate with multiple child service providers to establish the universal level of mental health services for our nation’s children, while supporting the primary goal of our educational institutions – to educate children. The next steps include adding prevention of mental health dysfunction to this reformed paradigm and expanding the services in the naturalistic environment, which, for children and adolescents, is schools.

REFERENCES Adelman, H. S., & Taylor, L. (1999). Mental health in schools and system restructuring. Clinical Psychology Review, 19, 137–163. American Psychiatric Association. (1999). Diagnostic and statistical manual – Fourth edition. Washington, DC: American Psychiatric Association. Blake, S. M., Ledsky, R., Goodenow, C., & O’Donnell, L. (2001). Receipt of school health education and school health service among adolescent immigrants in Massachusetts. Journal of School Health, 71, 105–113. Bloom, M. (1996). Primary prevention practices. New York: Sage Publications. Bradley, R., Henderson, K., & Monfore, D. A. (2004). A national perspective on children with emotional disorders. Behavioral Disorders, 29, 211–223.

School-Based Mental Health Services

49

Bradsher, K. (October 23, 1996). Putting values in classroom, carefully. New York Times. Retrieved August 6, 2005, from http://personal.ecu.edu/estesst/2323/Readings/ charactereducation.html Burns, B. J., Hoagwood, K., & Mrazek, P. J. (1999). Effective treatment for mental disorders in children and adolescents. Clinical Child and Family Psychology Review, 2, 199–254. Conroy, M. A., & Brown, W. H. (2004). Early identiﬁcation, prevention, and early intervention with young children at risk for emotional or behavioral disorders: Issues, trends, and a call for action. Behavioral Disorders, 29, 224–236. Consortium on the School-Based Promotion of Social Competence. (1996). The school-based promotion of social competence: Theory, research, practice, and policy. In: R. J. Haggerty, L. R. Sherrod, N. Garmezy & M. Rutter (Eds), Stress, risk, and resilience in children and adolescents (pp. 268–316). Cambridge, UK: Cambridge University Press. Council for Exceptional Children. (n.d.). IDEA law and resources. Retrieved June 7, 2005, from http://www.cec.sped.org/law_res/doc/law/law/index.php Davis, A. S., McIntosh, D. E., Phelps, L., & Kehle, T. J. (2004). Addressing the shortage of school psychologists: A summative overview. Psychology in the Schools, 41, 489–495. Doll, B. (August, 1999). Making schools resilient: A case for social environments as therapy. A Division 16 Presidential Address delivered at the annual convention of the American Psychological Association, Boston, MA. Eber, L., Sugai, G., Smith, C. R., & Scott, T. M. (2002). Wraparound and positive behavioral interventions and supports in the schools. Journal of Emotional and Behavioral Disorders, 10, 171–180. Elias, M. J., Zins, J. E., Graczyk, P. A., & Weissberg, R. P. (2003). Implementation, sustainability, and scaling up of social–emotional and academic innovations in public schools. School Psychology Review, 32, 303–319. Goldston, S. E. (1985). Primary prevention: Historical perspectives and a blueprint for action. American Psychologist, 41, 453–460. Greenberg, M. T., Domitrovich, C., & Bumbarger, B. (2001). The prevention of mental disorders in school-aged children: Current state of the ﬁeld. Prevention and Treatment, 4, Article 1. Retrieved April 16, 2005, from http://journals.apa.org/prevention/volume4/ pre0040001a.html Greenberg, M. T., Weissberg, R. P., O’Brien, M. U., Zins, J. E., Fredericks, L., Resnik, H., & Elias, M. (2003). Enhancing school-based prevention and youth development through coordinated social, emotional, and academic learning. American Psychologist, 58, 466–474. Hawton, K., Rodham, K., Evans, E., & Weatherall, R. (2002). Deliberate self harm in adolescents: Self report survey in schools in England. British Medical Journal, 325, 1207–1211. Hochschild, J., & Scovronick, N. (2003). The American dream and the public schools. Oxford, UK: Oxford University Press. Johnson, J. P., Schwartz, R. A., Livingston, M., & Slate, J. R. (2000). What makes a good elementary school? A critical examination. Journal of Educational Research, 93, 339–348. Kendziora, K., Bruns, E., Osher, D., Pacchiano, D., & Mejia, B. (2001). Systems of care: Promising practices in children’s mental health, 2001 series (Vol. 1). Washington, DC: Center for Effective Collaboration and Practice, American Institutes for Research. Kolbe, L. J., Collins, J., & Cortese, P. (1997). Building the capacity of schools to improve the health of the nation: A call for assistance from psychologists. American Psychologist, 52, 256–265.

50

MELISSA PEARROW AND PETER WHELLEY

Nastasi, B. K., Plymert, K., Varjas, K., & Moore, R. B. (2002). Exemplary mental health programs: School psychologists as mental health service providers (3rd ed). Washington, DC: National Association of School Psychologists. Nelson, G., Westhues, A., & MacLeod, J. (2003). A meta-analysis of longitudinal research on preschool prevention programs for children. Prevention and Treatment, 6, 6–31. New Freedom Commission on Mental Health. (2003). Achieving the promise: Transforming mental health care in America. Final Paper. DHHS Publication No. SMA-03-3832. Rockville, MD: U.S. Department of Health and Human Services. Olympia, D., Farley, M., Christiansen, E., Pettersson, H., Jenson, W., & Clark, E. (2004). Social maladjustment and students with behavioral and emotional disorders: Revisiting basic assumptions and assessment issues. Psychology in the Schools, 41, 835–847. Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American Psychologist, 44, 329–335. Petersen, G. J., Pietrzak, D., & Speaker, K. M. (1998). The enemy within: A national study on school violence and prevention. Urban Education, 33, 331–359. Pulliam, J. D., & Van Patten, J. (1995). History of education in America (6th ed). Englewood Cliffs, NJ: Prentice Hall, Inc. Reid, R., Gonzalez, J. E., Nordness, P. D., Trout, A., & Epstein, M. H. (2004). A meta-analysis of the academic status of students with emotional/behavioral disturbance. The Journal of Special Education, 38, 130–143. Rush, S., & Vitale, P. A. (1994). Analysis for determining factors that place elementary students at risk. Journal of Educational Research, 87, 325–333. Samaraweera, S., & Hurwitz, I. (2000). Primary prevention: Making it work. Chestnut Hill, MA: Life Studies Foundation, Inc. Sugai, G., Sprague, J. R., Horner, R. H., & Walker, H. M. (2000). Preventing school violence: The use of ofﬁce referrals to assess and monitor school-wide discipline interventions. Journal of Emotional and Behavioral Disorders, 8, 94–101. Taub, J., & Pearrow, M. (2005a). Resilience through violence prevention in schools. In: S. Goldstein & R. B. Brooks (Eds), Handbook of resilience in children (pp. 357–371). New York: Kluwer Publishers. Taub, J., & Pearrow, M. (August 2005b). Relationships between community-based wraparound services and children’s school functioning. Poster presentation at the annual conference of the American Psychological Association, Washington, DC. U.S. Department of Education. (n.d.a). History of the IDEA: Twenty-five years of progress in educating children with disabilities through IDEA. Retrieved June 4, 2005, from http:// www.ed.gov/print/policy/speced/leg/idea/history.html U.S. Department of Education. (n.d.b). Office of Special Education Office of Technical Assistance on Positive Behavioral Interventions and Supports: School wide PBS. Retrieved May 1, 2005, from http://www.pbis.org/schoolwide.htm U. S. Department of Education. (n.d.c). Ten facts about K-12 education funding. Retrieved April 23, 2005, from http://www.ed.gov/about/overview/fed/10facts/index.html#chart6 U.S. Department of Education, Ofﬁce of Special Education Programs, Data Analysis System (DANS). (2004). Table AA10: Percentage (based on 2003 population estimates) of children served under IDEA, part B by age group, in 2003. Retrieved April 18, 2005 from, http://www.ideadata.org/tables27th%5Car_aa10.htm

School-Based Mental Health Services

51

U.S. Department of Health and Human Services. (1999). Mental health: A report of the surgeon general. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health. U.S. Department of Health and Human Services. (2001a). Mental health: Culture, race, and ethnicity – A supplement to mental health: A report of the surgeon general – Executive summary. Rockville, MD: Department of Health and Human Services, U.S. Public Health Service, Ofﬁce of the Surgeon General. U.S. Department of Health and Human Services. (2001b). Youth violence: A report of the surgeon general – Executive summary. Rockville, MD: Department of Health and Human Services, U.S. Public Health Service, Ofﬁce of the Surgeon General. Wagner, M., Kutash, K., Duchnowski, A. J., Epstein, M. H., & Sumi, W. C. (2005). The children and youth we serve: A national picture of the characteristics students with emotional disturbances receiving special education. Journal of Emotional and Behavioral Disorders, 13, 79–96. Ward, J., & Losen, D. J. (2003). New directions for youth development – theory practice research: Deconstructing the school-to-prison pipeline. New York: Jossey-Bass Publishers/Wiley & Sons, Inc. Weissberg, R. P., Caplan, M., & Harwood, R. L. (1991). Promoting competent young people in competence-enhancing environments: A systems-based perspective on primary prevention. Journal of Consulting and Clinical Psychology, 59, 830–841. Weissberg, R. P., Caplan, M. Z., & Sivo, P. J. (1989). Promoting competent-enhancing environments: A systems-based perspective on primary prevention. Journal of Consulting and Clinical Psychology, 59, 830–841. Weist, M. D., & Evans, S. W. (2005). Expanded school mental health: Challenges and opportunities in an emerging ﬁeld. Journal of Youth and Adolescence, 34, 3–6. Winslow, E. B., Sandler, I. N., & Wolchik, S. A. (2005). Building resilience in all children. In: S. Goldstein & R. B. Brooks (Eds), Handbook of resilience in children (pp. 337–356). New York: Kluwer Publishers. Worthington, J., Hernandez, M., Friedman, B., & Uzzell, D. (2001). Systems of care: Promising practices in children’s mental health, 2001 series (Vol. II). Washington, DC: Center for Effective Collaboration and Practice, American Institutes for Research.

THE GREAT DIVIDE: HOW MENTAL HEALTH POLICY FAILS YOUNG ADULTS Maryann Davis and Nancy Koroloff INTRODUCTION All individuals are challenged by the movement from being an adolescent living at home and attending school to being an adult typically heading a household and working to support him or herself. This period of time is called the transition to adulthood and is even more challenging for youth from vulnerable populations such as youth with disabilities, in foster care, in juvenile justice system, and the like (Osgood, Foster, Flanagan, & Ruth, 2005). The ages that transition encompasses have not gained consensus in research literature or policy. It begins at ages 14–16 in policies such as the Individuals with Disabilities Education Act (IDEA; PL101-476, 1997 and 2004 amendments) or Federal programs such as the Social Security Administration’s SSI Youth Transition Demonstration Projects, which identiﬁes ages 22 and 25, respectively, as ending transition. Recent studies on young adulthood in the general population (Settersten, Frustenberg, & Rumbaut, 2005), found that by age 30, the rapid changes of young adulthood had typically stabilized. Thus, using the broadest ages indicated by policy and research, transition to stable adulthood encompasses ages 14–30.

Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 53–74 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14004-0

53

54

MARYANN DAVIS AND NANCY KOROLOFF

Youth with disabilities struggle to attain adult functioning (Wells, Sandefur, & Hogan, 2003; Blackorby & Wagner, 1996; Browning, Dunn, Rabren, Whetstone, & Mabrey, 1995). Students with serious emotional disturbance (SED) fare particularly poorly as they cross the threshold into adulthood compared with other students with disabilities (Blackorby & Wagner, 1996). Longitudinal studies have demonstrated young adult underemployment, homelessness, trouble with the law, youthful pregnancy, high school dropout, and co-morbid substance abuse or dependence among adolescents with SED (e.g., Vander Stoep et al., 2000; Davis, Williams, z& Fernandes, 2005; Armstrong, Dedrick, & Greenbaum, 2003; Davis & Vander Stoep, 1997). Most recently, comparing transition outcomes of secondary students with SED in 2001 to those in 1987 revealed that the poor dropout rate has not changed, students with SED were the only disability group to experience a signiﬁcant decline in employment, and the most concerning outcomes, such as school expulsion/suspension, being ﬁred from a job, or being arrested had actually worsened by 57% (Wagner, Cameto, & Newman, 2003). Impaired young adult functioning has been found in youth with SED in mental health and special education systems (e.g., Armstrong et al., 2003; Wagner, 1995) as well as in youth with psychiatric disorders in community-based samples (Davis et al., 2005; Vander Stoep et al., 2000). Thus, it is likely that adult functioning is impaired in most individuals who had SED as adolescents, estimated to be about 1.5–2 million individuals.1 Longitudinal studies of psychiatric disorders also indicate that these disorders are persistent into adulthood. Adolescents who have attention-deﬁcit/ hyperactivity disorder (Ingram, Hechtman, & Morgenstern, 1999), schizophrenia (Hollis, 2000), major depressive disorder (Lewinsohn, Rohde, Klein, & Seeley, 1999; Rao et al., 1995; Bardone, Mofﬁtt, Caspi, & Dickson, 1996), simple phobia (Pine, Cohen, Gurley, Brook, & Ma, 1998), or conduct disorder (Bardone et al., 1996), are likely to have the same disorder in young adulthood. Studies have also found that adolescent disorders, including anxiety disorders, affective disorders, and conduct disorders, are strongly predictive of other disorders in young adulthood (e.g., Kasen et al., 2001; Peterson, Pine, Cohen, & Brook, 2001; Pine et al., 1998; Biederman, Faraone, & Kiely, 1996; Bardone et al., 1996; Rao et al., 1995; Pollack et al., 1990; Pollack, Otto, Rosenbaum, & Sachs, 1992). The combination of poor young adult functioning with the continued presence of psychiatric disorders calls for examination of public policy and the way we organize services that would lead to improved outcomes in this population.

The Great Divide: How Mental Health Policy Fails Young Adults

55

EFFECT OF AGE CATEGORIES It could be argued that organizing mental health services into separate child and adult systems does not serve the transition-aged population well. In all but one state, the child mental health and adult mental health systems at the state level are largely administratively separate, and often with very separate funding streams (Davis, 2003; Davis, Yelton, Katz-Leavy, & Lourie, 1995). The upper age limit for the child mental health system is typically age 18, though it is age 21 in about a quarter of states (Davis & Sondheimer, 2005). Adult mental health systems serve individuals who are 18 and older. Federal policies and programs mirror this same categorization scheme. Judge David L. Bazelon Center for Mental health Law (2005) reviewed 57 federal programs that affect transition aged youth with serious mental health conditions. These programs were generated from federal policy. Of these 57, 22 limit the services provided through the program to youth variously deﬁned as those up to ages 18, 19, 21, 23, or 25. Two of the larger programs, Medicaid and Supplemental Security Income (SSI), offer beneﬁts to all ages but change the rules at a certain age. Medicaid rules change at ages 18, 19, 20, or 21 depending on the state. SSI rules change at age 18. Because our services and policies are divided into child and adult categories, as youth mature across the adult threshold they are likely to encounter challenges to accessing needed services caused by change in their chronological age but not change in need. The target population deﬁnition that shapes access to the $436.1 million dollar block grant (FY05) distribution from the Center for Mental health Services (CMHS) of the federal Substance Abuse and Mental health services Administration is ‘‘children with serious emotional disturbance’’ and ‘‘adults with a serious mental illness’’ as deﬁned in the Federal Register (FR, 58(96). P. 29422). The block grants provide funds to states for the provision of community mental health services. Thus, the development of services within state mental health is shaped by these federal priority population deﬁnitions. As can be seen from the above description of state mental health organization and federal policies, access to services and programs is limited by ages largely deﬁned by child or adult status. The formal policy that describes this is eligibility criteria or target population deﬁnition. Eligibility criteria refers to a series of conditions that individuals must have in order to access services. Target or priority population definitions, typically lay out a series of conditions that deﬁne who in the population has priority in accessing services. While a large number of individuals may meet the eligibility criteria, a much smaller group will be included in the target or priority population.

56

MARYANN DAVIS AND NANCY KOROLOFF

One of the results of prioritizing populations, giving them preferential access to services and designing services for them, is that non-priority groups experience denial of services, access to a very limited array of services, or access to less than appropriate services. State child mental health administrators overwhelmingly report that eligibility criteria or target population deﬁnitions are different for child and adult mental health systems, and that the differences produce a barrier to the continuation of needed services and supports for at least some youth once they reach the upper age limit for the child system (Davis & Sondheimer, 2005). The general opinion among these administrators was that the adult system deﬁnitions were narrower than the child system deﬁnitions, and speciﬁcally precluded ‘‘child’’ diagnoses. Foremost among these would be diagnoses such as attention deﬁcit disorder, hyperactivity, conduct disorder, and oppositional deﬁant disorder (Davis, 2001). If eligibility criteria or target population deﬁnitions differ between the child and adult mental health systems then access to needed services during the transition to adulthood becomes arbitrary based on change in age, not change in need. The present study describes age-based deﬁnitional agreement and disagreement at the state and federal level: (1) between the federal deﬁnitions of SED and serious mental illness (SMI) put forth by the CMHS and (2) between state population deﬁnitions for adult and child mental health systems. Through analysis of federal and state-level policies, we explore the impact of age-based population policies on care continuity for youth from the child system as they age into adulthood.

METHODS Data The present study was initiated as part of a larger study to determine the status of transition support development in state child mental health systems and is described in Davis and Sondheimer (2005). Brieﬂy, the National Association of State Mental health Program Directors (NASMHPD) provided a list of the lead child mental health administrator in each state and the District of Columbia in 2001. Each member was sent a cover letter from the ﬁrst author and from an administrator at NASMHPD’s National Technical Assistance Center describing the purpose and nature of the study and encouraging participation. Each member was contacted and completed a semi-structured interview. In part of the interview administrators were asked

The Great Divide: How Mental Health Policy Fails Young Adults

57

about any policies that impacted the transitioning population. Those policies were described during the interview and copies of them were requested. Administrators were also asked to send eligibility criteria for the child and adult mental health system in their state. Within the ‘‘eligibility’’ policies, some states indicated they did not have eligibility criteria, but rather used target or priority population deﬁnitions to deﬁne their population. When administrators were asked to update those policies in June 2003 for purposes of this analysis, they were asked to send eligibility criteria or target population deﬁnitions, leaving it up to the respondent to choose which to send. For brevity, then, when not distinguishing between eligibility criteria and target population deﬁnitions, both are referred to herein as ‘‘population policies’’. Responses were obtained from 48 states, 46 of which made both the child and adult policy available, including the District of Columbia. The state-level analyses below are based on those 46 states (and do not reﬂect the policies of AL, ND, NM, RI, or SC). The federal Center for Mental health Services’ deﬁnitions of SED and SMI (FR, 58(96). P. 29422), were used as the basis for analyzing federal level policy. The ﬁrst author developed a coding scheme for the content of the state criteria/deﬁnition, using a loose grounded theory coding scheme (Glaser & Strauss, 1967). Beginning with the ﬁrst population policy, each concept it contained was deﬁned and labeled (i.e., age covered, presence of a diagnostic component, and presence of a functional component). That coding scheme was then applied to the second population policy. If the second population policy introduced new concepts, they were deﬁned and labeled. If the second population policy required that previous concepts be reﬁned then the ﬁrst policy was recoded in accordance. This coding scheme was then applied to the third population policy. This iterative process was applied until there were no new concepts that emerged and all ﬁnal concepts had been coded for each policy. Population Policy Coding Scheme No state declared that they had one population policy for all, and each policy came labeled as a child or adult policy. Table 1 presents the coding scheme for child and adult population policies It should be noted that the dimension of a ﬁscal requirement could only be coded if it was articulated within the submitted population policy. It is possible that ﬁscal criteria were stated elsewhere and expected to be implicit in the policy sent. For example, if a state’s mental health authority only served clients who are eligible for Medicaid, this criteria may have been such a pervasive

Discovered Dimensions in Child and Adult Population Policies.

Concept Child Population Policies Age at which child services end Requires a DSM-III-R or DSM-IV diagnosis (or ICD equivalent)

Deﬁnition of functional impairment Impairment duration requirement Fiscal requirement

Other qualifying conditions

Codes

18/19/21/21 if enter by 18 Yes/No/Diagnosis is one of more than one qualifying condition (at least one of which is required) Yes/No

Yes/No

Yes/No/Functional impairment is one of more than one qualifying condition (at least one of which is required) Cutoff score on scale/Meets description of area and level of impairment None/Lasted 6 mo/Expected to last 6 mo/Lasted 1 year/Expected to last 1 year None/235–200–150% of federal poverty level/Has Medicaid, no insurance, or used up resources/No Medicaid/Uninsured and indigent/Fiscal requirement is one of more than one qualifying condition (at least one of which is required) Risk of out-of-home placement/Multiagency or interdisciplinary team involvement/History of intensive or out of home services/ Presence or risk of psychosis, suicide, dangerous to others/ Homeless and emotionally disturbed/‘‘Other’’

MARYANN DAVIS AND NANCY KOROLOFF

The presence of a list of qualifying diagnoses (not including the usual exclusions of ‘‘V’’ codes, substance use disorders, or developmental disabilities unless co-occurring with qualifying diagnosis) When diagnoses were required or qualifying, whether the following diagnoses were included; psychotic disorders, major affective disorders, borderline personality disorder, post traumatic stress disorder, attention deﬁcit and disruptive behavior disorders Requirement of functional impairment

58

Table 1.

The presence of a list of qualifying diagnoses (not including the usual exclusions of ‘‘V’’ codes, substance use disorders, or developmental disabilities unless co-occurring with qualifying diagnosis) When diagnoses were required or qualifying, whether the following diagnoses were included; psychotic disorders, major affective disorders, borderline personality disorder, post traumatic stress disorder, attention deﬁcit and disruptive behavior disorders Requirement of functional impairment When functional impairment was required or qualifying; deﬁnition of functional impairment When functional impairment required or qualifying; impairment duration requirement Fiscal requirement

Other qualifying conditions

18/19 Yes/No/Diagnosis is one of more than one qualifying condition (at least one of which is required) Yes/No

Yes/No

Yes/No/Functional impairment is one of more than one qualifying condition (at least one of which is required) Cutoff score on scale/Meets description of area and level of impairment None/Lasted 6 mo/Expected to last 6 mo/Lasted 1 year/Expected to last 1 year/Lasted 2 years/Duration or other dimension (i.e., intensity) qualifying None/Fiscal requirement is one of more than one qualifying condition (at least one of which is required)/200% federal poverty level and Not Medicaid/150% federal poverty level/ 100% of federal poverty level/No Medicaid/Indigent/Risk of dependence/Social security disability income recipient/Unknown poverty level History of intensive or out of home services/Presence or risk of psychosis, suicide, dangerous to others/Disaster or crisis exposure/Homeless and mentally ill/Arrested or convicted/ ‘‘Other’’

The Great Divide: How Mental Health Policy Fails Young Adults

Adult Population Policies Age at which adult services can begin Requires a DSM-III-R or DSM-IV diagnosis (or ICD equivalent)

59

60

MARYANN DAVIS AND NANCY KOROLOFF

characteristic of care that it was not speciﬁcally mentioned in eligibility or priority population deﬁnitions. While ﬁscal characteristics that were speciﬁed in population policies are summarized in Table 2 below, no further analysis of this dimension is offered because of the likelihood of missing information. Priority Population versus Eligibility Criteria Administrators were not asked to clarify whether the population policy they were sending was an eligibility criteria or a target/priority population deﬁnition. The request simply asked for ‘‘Eligibility criteria/target population deﬁnition for child mental health (or adult mental health)’’. Twenty-six states sent policies labeled ‘‘eligibility’’ criteria for both child and adult services, six states sent policies labeled ‘‘target’’ or ‘‘priority’’ population for both child and adult services, and four states sent either mixed or unclear policies (e.g., eligibility criteria for one age group and an unlabeled policy for the other age group). For the remaining 10 states the type of population policy sent was unclear. In addition, two of the eligibility criteria were for speciﬁc services, such as ‘‘community support programs for chronically mentally ill persons’’ that appeared to be a specialized subset of available services, and it was unclear whether that criteria was applied to all services or just the speciﬁed one. Other states (n ¼ 7) sent multiple eligibility criteria that were for various levels of care. For example, the criteria for outpatient services, for emergency/crisis services, and for intensive case management services may have all differed. When multiple criteria were sent, the criteria for the most intensive child and most intensive adult service were used for comparison.

RESULTS Federal Definitions The following are the CMHS deﬁnitions of SED and SMI. Children with Serious Emotional Disturbance (p. 29425) ‘‘children with serious emotional disturbance’’ are persons from birth to age 18, who currently or at any time during the past year, have had a diagnosable mental, behavioral, or emotional disorder of sufﬁcient duration to meet diagnostic criteria speciﬁed within DSM-III-R that has resulted in functional impairment, which substantially interferes with or limits the child’s role or functioning in family, school, or community activities.

Frequency of Dimensions of Adult and Child Population Policies.

Concept

Age at which child services end and adult service begin (Child N ¼ 45, Adult N ¼ 44)

Requires a DSM-III-R/-IV diagnosis, or ICD equivalent (N ¼ 46) Included diagnoses when diagnosis a qualifying condition (Child N ¼ 38, Adult N ¼ 44)

Requirement of functional impairment (N ¼ 46)

Deﬁnition of functional impairment when it was qualifying conditionb (Child N ¼ 43, Adult N ¼ 40) Impairment duration requirement when functional impairment a qualifying condition (Child N ¼ 43, Adult N ¼ 40)

Value

Age 18 Age 19 Age 21 Ends at age 21 if entered oage 18 Age 22 Ends at age 22 if entered oage 18 Yes No This or other conditions qualifya Psychotic disorders Major affective disorders Borderline personality disorder Post traumatic stress disorder Attention deﬁcit/disruptive behavior disorders Yes No This or other conditions qualifya Description of areas and impairment levels Cutoff score on scale

Child

Adult

68.9 2.2 11.1 13.3 2.2 2.2 76.1 17.3 6.5 100.0 100.0 100.0 92.1 97.4 63.0 6.5 30.4 76.7 25.6

100.0 0 0 NA 0 NA 93.5 4.3 2.2 100.0 100.0 76.7 65.1 39.5 78.3 6.5 15.2 79.1 20.9

53.5 7.0 2.3 9.3 20.9 0

45.0 5.0 2.5 12.5 20.0 12.5

61

None Lasted 6 mo Expected to last 6 mo Last 1 year Expected to last 1 year Lasted 2 years

% State Policies

The Great Divide: How Mental Health Policy Fails Young Adults

Table 2.

62

Table 2. (Continued ) Concept

Value

% State Policies

a

Other qualifying conditions (N ¼ 46)

Adult

7.0 28.3

2.5 37.0

21.7 19.6 8.7 0.0 4.3 34.8

13.0 0.0 0.0 8.7 10.9 23.9

Policy stipulates that this condition qualiﬁes, but is not required if other conditions are met. Some states required both areas of signiﬁcant functional impairment and a cutoff score on a scale of functional impairment, thus columns do not sum to the total N.

b

MARYANN DAVIS AND NANCY KOROLOFF

a

This or other conditions qualify Risk or history of out-of-home placement or other intensive services Presence/risk psychosis/dangerous to self/others Multiagency/interdisciplinary team involvement Special education student Arrested/convicted of crime Homeless and mentally ill Other

Child

The Great Divide: How Mental Health Policy Fails Young Adults

63

These disorders include any mental disorder (including those of biological etiology) listed in DSM-III-R or their ICD-9-CM equivalent (and subsequent revisions) with the exception of DSM-III-R ‘V’ codes, substance use, and developmental disorders, which are excluded, unless they co-occur with another diagnosable serious emotional disturbance. All of these disorders have episodic, recurrent, or persistent features; however, they vary in terms of severity or disabling effects. yFunctional impairment is deﬁned as difﬁculties that substantially interfere with or limit a child or adolescent from achieving or maintaining one or more developmentally appropriate social, behavioral, cognitive, communicative, or adaptive skills. Functional impairments of episodic, recurrent, and continuous duration are included unless they are temporary and expected responses to stressful events in their environment. Children who would have met functional impairment criteria during the referenced year without the beneﬁt of treatment or other support services are included in this deﬁnition.

Adults with Serious Mental Illness. (p. 29425) ‘‘adults with serious mental illness’’ are persons age 18 and over, who currently or at any time during the past year, have had a diagnosable mental, behavioral, or emotional disorder of sufﬁcient duration to meet diagnostic criteria speciﬁed within DSM-III-R that has resulted in functional impairment, which substantially interferes with or limits one or more major life activities. These disorders include any mental disorder (including those of biological etiology) listed in DSM-III-R or their ICD-9-CM equivalent (and subsequent revisions), with the exception of DSM-III-R ‘‘V’’ codes, substance use disorders, and developmental disorders, which are excluded, unless they co-occur with another diagnosable serious mental illness. All of these disorders have episodic, recurrent, or persistent features; however, they vary in terms of severity or disabling effects. yFunctional impairment is deﬁned as difﬁculties that substantially interfere with or limit role functioning in one or more major life activities including basic daily living skills (e.g., eating, bathing, dressing); instrumental living skills (e.g., maintaining a household, managing money, getting around the community, taking prescribed medication); and functioning in social, family, and vocational/educational contexts. Adults who would have met functional impairment criteria during the referenced year without beneﬁt of treatment or other support services are considered to have serious mental illnesses.

The language of these two deﬁnitions is remarkably similar. There is no explicit difference in the diagnostic criteria referenced in the two deﬁnitions. Implicit differences could, however, be generated from some diagnostic categories. For example, antisocial personality disorder (APD), a diagnosis which cannot be assigned prior to the age of 18, implicitly restricts APD from any list of qualifying child diagnoses, and makes the diagnostic portion of the CMHS SED and SMI deﬁnitions implicitly different. Fortunately, the DSM-IV has few diagnostic categories that would lead to further implicit differences in the diagnostic portion of the two deﬁnitions. Many diagnoses

64

MARYANN DAVIS AND NANCY KOROLOFF

require onset by a certain age. For example, separation anxiety disorder must ensue before age 18. However, if the age of onset is appropriate, most diagnostic categories can be assigned across the life span. Many disorders included in the CMHS deﬁnitions occur more commonly at earlier or later ages, but are not precluded in other ages. For example, personality disorder diagnoses are generally made on the basis of consistent behaviors and traits characteristic of a person’s functioning ‘‘since early adulthood’’ (p. 335). However, with the exception of APD, any personality disorder can be diagnosed in adolescence when the ‘‘maladaptive traits appear to be stable’’ (p. 336), and the adolescent meets the diagnostic criteria. Conversely, conduct disorder, which is more common in adolescents, can be applied to adults who do not meet criteria for APD. Thus, with the exception of APD, there is no diagnostic difference in the CMHS SED and SMI deﬁnitions. Aside from the age difference in the two CMHS deﬁnitions, the only remaining difference in terminology is in the deﬁnitions of ‘‘functional impairment’’. For those under age 18 the focus of functional impairments is on age-appropriate skills, whereas for those over 18 functioning is not tied to developmental considerations. Thus, ‘‘role functioning in one or more major life activities’’ would apply to anyone 18 or over. In applying the Adult with Serious Mental Illness deﬁnition to an 18-year-old it might be difﬁcult to determine whether their inability to maintain a household or manage money is related to their relative immaturity or to their psychiatric condition. A 17.9-year-old may meet the child functional impairment criteria because of their failure to progress on developmental tasks (e.g., behaviorally or cognitively immature) but would only qualify for the adult deﬁnition (0.1 years later) if that immaturity substantially prohibits him or her from achieving basic daily living skills, instrumental living skills, or functioning in social, family, and vocational/educational contexts. While it is likely that impairment that ‘‘substantially’’ interferes with the achievement of age-appropriate cognitive or behavioral functioning in an 18-year-old would also substantially interfere with one of the areas of adult functioning, it is easily possible that an individual deﬁned as having SED might not meet the functional criteria of having SMI on their 18th birthday according to the CMHS deﬁnitions.

State Definitions Ages Served As can be seen from Table 1, state population policies varied along several dimensions. The distribution of each policy dimension can be seen in Table 2.

The Great Divide: How Mental Health Policy Fails Young Adults

65

The upper age limit for the majority of state child systems is 18, with a smaller number continuing to age 21, and 3 states ending at 19 or 22. For most of the states that served youth up to age 21 or 22 in the child system, access to that system after age 18 was only available to those who were already in the system when they turned 18. All state adult systems served individuals aged 18 and older. Diagnostic Comparability Most states included some aspects of diagnosis in their child or adult population policy. As can be seen from Table 2, for a small proportion of states, diagnoses were not required if other conditions were met (such as having recently attempted suicide). More states require the presence of a diagnosis for the adult population than for the child population. Within policies that included diagnoses as qualifying conditions, adult population policies were less likely to include post traumatic stress disorder, borderline personality disorder, or attention deﬁcit or disruptive behavior disorders as qualifying diagnoses than child population policies. Child and adult policy diagnostic requirement were compared within states. Child and adult diagnostic requirements were considered to be the same if they used the same language (which, like the federal deﬁnitions, could include an implicit difference), or had the same list of qualifying diagnoses, or had the same absence of diagnostic requirement. This analysis revealed that 23.9% of states have the same adult and child diagnostic requirements, and an additional 4.3% had no speciﬁc diagnostic requirement for either their child or adult population policy (e.g., broadly requiring a ‘‘mental disorder’’, but no requirement to meet diagnostic criteria). In 47.8% of states both child and adult population policies required diagnoses but the adult requirement was more restrictive by virtue of fewer speciﬁc diagnoses meeting the criteria (e.g., the child deﬁnition included any DSMIV disorder and the adult deﬁnition had a speciﬁc list of qualifying diagnoses, or both the child and adult system had lists, but the adult list was shorter and there were qualifying child diagnoses that did not meet adult criteria). In 15.2% of states there was no diagnostic requirement in the child population policy and some diagnostic requirement in the adult policy, and in 6.5% of states adult and child policies had lists of qualifying diagnoses that were simply different (not more or less restrictive compared with the other). In one state, diagnoses were required for the child deﬁnition, but in the adult deﬁnition diagnosed conditions could qualify adults, but other conditions could also. Overall then, based on diagnoses, adult and child policies were different in 69.6% of states, and in 90.6% of those the

66

MARYANN DAVIS AND NANCY KOROLOFF

diagnostic criteria were more restrictive for the adult policy. Child and adult policies that were diagnostically the same primarily used diagnostic language similar to the Center for Mental health Services’ target population deﬁnitions described above. Functional Comparability Population policies varied as to whether or not a functional impairment was required (see Table 2). Functional impairment was usually expressed in terms of the way the disability affected one or more domain of daily life. For children, functional domains often referred to capacity to attend school or live at home. For adults, functional domains included the ability to maintain employment or perform skills of daily living, such as bathing, cooking and the like. Some policies explicitly required functional impairment, some policies had no functional impairment in their policy, and others described functional impairment as being one of several conditions that could qualify the individual as part of the population. For example, in several states, for those children who met diagnostic criteria, the additional condition of functional impairment, out-of-home treatment, or suicide attempt each would qualify a child as being part of the target population. Comparing the child and adult functional impairment policy within states (N ¼ 46) revealed that in 65.2% of states both the adult and child policy were the same in the degree to which functional impairment was a criteria (i.e., it was required, or it was one qualifying condition in both the child and adult policy). In 21.7% of states the adult policy required functional impairment while the child policy either did not have a functional requirement or it was one possible qualifying condition. In 10.9% of states the reverse was true, and in one state, functional impairment was one qualifying condition in the child policy and not included in the adult policy. Policies that included consideration of functional impairment often contained requirements about the duration of the functional impairment (46.5% of child policies and 55% of adult policies; see Table 2). Within the 67.4% of states that included a functional impairment condition in both the child and adult policy, 58.1% had the same duration requirement, 22.6% had a more restrictive adult duration, and 19.4% had a more restrictive child duration. Overall, considering both degree and duration of functional impairment, the child and adult population policies were comparable in 39.1% of states. In addition to the degree to which functional impairment was a requirement, and the duration of the requirement, there was speciﬁc language deﬁning functional impairment as impaired domains of living (such as working,

The Great Divide: How Mental Health Policy Fails Young Adults

67

schooling, living with family), or as a score on a functional scale (see Table 2). Within the 39.1% of states in which the child and adult population policies were comparable in the degree and duration of the requirement, in only one state was the speciﬁc language describing functional impairment the same in the child and adult policy (WA). That state had distinctly different diagnostic requirements in the child and adult policy. Thus, considering just the diagnostic and functional impairment portions of population policies, no state had the same child and adult population policies. Additional dimensions were found in child and adult policies (see Table 2). Adult policies more typically included consideration of homelessness than child policies, whereas some child policies considered multiagency or interdisciplinary team involvement as qualifying conditions, while no adult policies included this dimension. Corrective Policies Administrators in three states (6.5%) indicated that they had policies that were aimed at correcting the inequalities of the child and adult population policies. This was achieved through automatic admission, or ‘‘grandfathering’’ of eligibility. In two states (MD, MA) this was applied to those youth who were receiving intensive services while approaching the upper age limit of child mental health services. These youths were automatically eligible for adult mental health services and did not need to complete the adult eligibility process. Oklahoma had the broadest grandfathering eligibility policy. In that state, an individual who qualiﬁed for state mental health services as a child or youth was automatically eligible for adult mental health services as long as they were ﬁnancially eligible. Financial eligibility could pose an arbitrary barrier since a young person could be determined ‘‘indigent’’ (a criteria for services) under the parents’ household income until age 21. At that point they must reestablish ﬁnancial eligibility based on their own income. Thus, these policies did not completely remove arbitrary barriers of access caused by population policies, but did remedy it for certain subgroups of youth.

DISCUSSION Methodological Limitations There were several limitations to the information used in the analysis in this article. As noted earlier, it is unclear whether states that sent target population deﬁnitions have eligibility criteria that parallel the target population deﬁnition, or have broader eligibility criteria. Further, not all policies were

68

MARYANN DAVIS AND NANCY KOROLOFF

labeled in a way that clariﬁed whether they represented eligibility criteria or target/priority population deﬁnitions. Thus, the extent to which the reviewed deﬁnitions translate into the potential direct denial of services, as would be with eligibility criteria, is a rough estimation. States were also not explicitly asked to provide information on ﬁnancial eligibility criteria although some states included that type of information in the policy they submitted. Thus, the current analysis does not fully address the role that ﬁnancial status may play in the alignment of adult and child population policies. Along a related vein, eligibility for Medicaid was part of the explicit criteria for at least two states, and like ﬁscal requirements, may have been an implicit requirement for other states. However, it was beyond the scope of this study to examine the myriad federal regulations on eligibility for Medicaid, and medically necessary services, as well as each state’s speciﬁc conﬁguration of Medicaid regulations. Further, this study does not address ‘‘on the street’’ implementation of these policies (Stowe & Turnbull, 2001, p. 210). As one administrator expressed, no doubt there is some ‘‘gaming’’ of the deﬁnitions to get access to services. Another concern may be that only one researcher coded the policies raising the possibility of a systematic bias in the coding. As is often true of policy research, not all states’ administrators agreed to participate in the study by providing the researchers with population policies from their states. Population policies from both child and adult mental health were received from 45 states and the District of Columbia, and represent nearly all of the states that could have responded. Thus, the data reported here provides a useful, if not complete, picture of the nation as a whole. Consequences of Age-Based Discrepancies Despite these limitations, this study clearly demonstrates pervasive agebased discrepancies in state level child and adult population policies. The nature of the policy discrepancy is idiosyncratic to the state an individual lives in. Discrepancies at the state level are generally greater than the discrepancies between the CMHS SED and SMI deﬁnitions at the federal level. It is likely that mental health population policies based on a false dichotomy between childhood and adulthood interfere with the continuity of care and with age-appropriate services during the transition from adolescence to adulthood. If the goal of services is to support individuals with serious mental health conditions to transition from adolescent to adult functioning, then these ﬁndings suggest that many youth encounter barriers to access because of age-based population policies.

The Great Divide: How Mental Health Policy Fails Young Adults

69

These deﬁnitional differences cited above highlight the difﬁculties of ‘‘institutional transitions’’. Institutional transitions occur when there is a change in status for individuals as they move from one institutional environment to another. Institutional transitions are mediated by bureaucratic and legal, rather than cultural or natural guidelines (Mallory, 1995). Institutional transitions are more arbitrary and rigid than ‘‘developmental transitions’’ which refers to the natural process of maturation and increased competence and the social changes that are associated with this natural process. One of the most arbitrary foundations of the population deﬁnitions reviewed is the notion that serious mental health conditions are different in adults and children, and that there is a speciﬁc and uniform age at which those differences apply. While it is reasonable to think that serious mental health conditions would have different manifestations at different developmental stages, and that some diagnoses are more common at one developmental stage than another, it is arbitrary to choose a certain birthday, such as 18 or 21, at which the meaning of the most serious mental health conditions uniformly change. Given the potentially negative consequences of the relatively arbitrary distinction between child and adult population policies, several remedies are offered. Remedies One approach to rectify age-based population policy barriers found in the reviewed state policies was the ‘‘grandfathering’’ of eligibility, described in the results. Grandfathering of eligibility for youth in the most restrictive settings provides an important bridge for continuity of services, but does nothing to remove barriers of access for young adults who are functioning in the community or in less restrictive settings. Structuring policy in this way could also lead to over-referring youth to more restrictive settings in order to increase the probability that they might receive adult services. Further, even grandfathering the entire served child mental health population for adult eligibility still maintains some implicit unfairness. If a young person has not been recognized as having a mental health need prior to their 18th birthday, but is recognized as such shortly after their 18th birthday, they will have to meet the more stringent adult criteria in order to access services, even though their condition could be just as disabling as a grandfathered youth who does not meet adult criteria. A more logical, fair, and developmentally guided approach would be to align the adult and child deﬁnitions by having the same diagnostic criteria and the functional criteria developmentally deﬁned. The federal CMHS

70

MARYANN DAVIS AND NANCY KOROLOFF

deﬁnitions serve as a good example. Their functional deﬁnition could be changed to: Functional impairment is deﬁned as difﬁculties that substantially interfere with or limit an individual from achieving or maintaining one or more developmentally appropriate social, behavioral, cognitive, communicative, or adaptive skills, or functioning in social, family, and vocational/educational contexts. Adaptive skills include self care, home living, community use, self-direction, health and safety, functional academics, and work. (Luckasson & Reeve, 2001)

With the substitution of these two sentences for the functional deﬁnition sentence in each of the two deﬁnitions, the two would line up and there would be no policy basis for individuals falling outside the target population as a consequence of reaching a particular birthday. One of the historical reasons for the CMHS SED deﬁnition was to recognize that children could have serious mental health conditions, and that their services needed to be different from that of adults (Knitzer, 1983). One of the risks of merging the two deﬁnitions would potentially be the loss of that recognition. However, much progress in epidemiology and mental health services research has occurred since 1983 and the presence of serious mental health conditions in children and adolescents is well documented. It seems unlikely that having parallel deﬁnitions would lead to the loss of appropriate services for children and adolescents. Another argument for constructing parallel deﬁnitions is to bring a stronger developmental framework to adult mental health services. As it stands, the current deﬁnition of SMI does not recognize developmentally based differences in functioning across the adult life span. Embracing this notion at the federal level could provide leadership in addressing developmentally different needs of younger, mature, and older adults. One of the consequences of aligning federal and state child and adult population deﬁnitions is that more individuals would be eligible for adult services than are currently. Financing strategies will need to be examined in order for broader eligibility to translate into more individuals accessing services. It would be unfortunate if aligning the deﬁnitions resulted in simply a different population going unserved in order to accommodate this ‘‘new’’ population.

CONCLUSIONS Why do we organize our policy and programs this way? According to cognitive psychologists, humans, like most animals, tend to group information

The Great Divide: How Mental Health Policy Fails Young Adults

71

into categories: is it edible or poisonous; male or female; homo or australopithecine; animal, vegetable, or mineral. The ability to rapidly categorize information promotes survival (Ashby & Maddox, 2005). In the world of scarce service resources, categorization helps providers make decisions despite the complex interactions between service priorities. When policies provide clear lines of demarcation, it is easier for the practitioner to make a swift and defendable decision in the face of many individuals whose needs will not be met. Policies that set up categories facilitate decision making, and solve problems (e.g., Lewandowsky, Kalish, & Ngang, 2002; Markman & Ross 2003). While categorization fulﬁlls positive functions, it also poses challenges when we examine complex phenomena, when categorization errors are made, or when applied to phenomena that are, in fact, continuous and not categorical. We have developed phrases about the conundrums of using categories; ‘‘it falls into a gray area’’, ‘‘he fell through the cracks’’ or ‘‘thinking outside of the box’’. For some aspects of our social system, categorization works relatively well (e.g., at 16 you can get a drivers license; at 18 you can vote). Although these strict age-based policies do not take into account individual developmental differences, this lack does not fundamentally change the outcome for the individual. The organization of mental health systems into child and adult serving systems is an example of categorizing a developmental process which is in fact a continuous phenomenon. The impact of this categorization scheme does fundamentally affect the lives of the young person who needs mental health services but is denied them because his age category supercedes his individual mental health needs.

NOTES 1. This ﬁgure was calculated by applying SED prevalence rates (5–9%, Costello, 1999; Friedman, Katz-Leavy, Manderscheid, & Sondheimer, 1996) to the 2000 U.S. Census estimates of 39 million 13–20 year olds would be 18–25 in 2005 (http://factﬁnder.census.gov/servlet/QTTable?_bm=y&-geo_id=01000US&-qr_name= DEC_2000_SF1_U_QTP2&-ds_name=DEC_2000_SF1_U&-_sse=on).

ACKNOWLEDGMENT This research was supported by a contract from the Center for Mental health Services, Substance Abuse and Mental Health Services Administration, through the American Institutes of Research (CMHS-0990-0115). The

72

MARYANN DAVIS AND NANCY KOROLOFF

authors would like to thank Janice Robert for her persistence in obtaining population policies, the mental health administrators that provided copies of population policies, and Krista Kutash and Steven Banks for comments on previous versions of this article.

REFERENCES Armstrong, K., Dedrick, R., & Greenbaum, P. (2003). Factors associated with community adjustment of young adults with serious emotional disturbance: A longitudinal analysis. Journal of Emotional and Behavioral Disorders, 11, 66–76. Ashby, F. G., & Maddox, W. T. (2005). Human category learning. Annual Review of Psychology, 56, 149–178. Bardone, A., Mofﬁtt, T., Caspi, A., & Dickson, N. (1996). Adult mental health and social outcomes of adolescent girls with depression and conduct disorder. Development and Psychopathology, 8(4), 811–829. Biederman, J., Faraone, S. V., & Kiely, K. (1996). Comorbidity in outcome of attention-deﬁcit/ hyperactivity disorder. In: L. Hechtman (Ed.), Do they grow out of it? Long-term outcomes of childhood disorders (pp. 39–75). Washington, DC: American Psychiatric Association. Blackorby, J., & Wagner, M. (1996). Longitudinal postschool outcomes of youth with disabilities: Findings from the National Longitudinal Transition Study. Exceptional Children, 62(5), 399–413. Browning, P., Dunn, C., Rabren, K., & Whetstone, M. (1995). Post-school outcomes for students with disabilities: A U.S. synopsis. Issues in Special Education and Rehabilitation, 10(1), 33–40. Costello, E. J. (1999). Prevalence and impact of parent-reported disabling mental health conditions among U. S. children: Comment. Journal of the American Academy of Child and Adolescent Psychiatry, 38(5), 610–613. Davis, M. (2001). Transition supports to help adolescents in mental health services. Alexandria, VA: National Association of State mental health Program Directors. Davis, M. (2003). Addressing the needs of youth in transition to adulthood. Administration and Policy in Mental Health, 30, 495–509. Davis, M., & Sondheimer, D. (2005). Child mental health systems’ efforts to support youth in transition to adulthood. Journal of Behavioral Health Services and Research, 32, 27–42. Davis, M., & Vander Stoep, A. (1997). The transition to adulthood for youth who have serious emotional disturbance: Developmental transition and young adult outcomes. The Journal of Mental Health Administration, 24(4), 400–427. Davis, M., Williams, V., & Fernandes, B. (2005). Substance abuse and the functioning of transition-aged youth with psychiatric disorders. The 17th annual research conference proceedings, A system of care for children’s mental health: Expanding the research base (March, 2004). Tampa, FL: University of South Florida, Louis de la Parte Florida Mental health Institute, Research and Training Center for Children’s Mental health (pp. 505–509). Davis, M., Yelton, S., Katz-Leavy, J., & Lourie, I. (1995). Unclaimed children revisited: The status of state children’s mental Health service systems. Journal of Mental health Administration, 22, 147–166.

The Great Divide: How Mental Health Policy Fails Young Adults

73

Friedman, R., Katz-Leavy, J., Manderscheid, R., & Sondheimer, D. (1996). Prevalence of serious emotional disturbance in children and adolescents. In: R. W. Manderscheid & M. A. Sonnenschein (Eds), Mental health, United States, 1996 (pp. 77–91). Washington, DC: U.S. Government Printing Ofﬁce. Glaser, B., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. New York: Aldine de Gruyter. Hollis, C. (2000). Adult outcomes of child- and adolescent-onset schizophrenia: Diagnostic stability and predictive validity. American Journal of Psychiatry, 157(10), 1652–1659. Ingram, S., Hechtman, L., & Morgenstern, G. (1999). Outcome issues in ADHD: Adolescent and adult long-term outcome. Mental Retardation and Developmental Disabilities Research Reviews, 5(3), 243–250. Judge David L. (2005). Bazelon Center for Mental Health Law. Moving On; Federal Programs to Assist Transition-Age Youth with Serious Mental Health Conditions. Washington, DC. Kasen, S., Cohen, P., Skodol, A. E., Johnson, J. G., Smailes, E., & Brook, J. S. (2001). Childhood depression and adult personality disorder: Alternative pathways of continuity. Archives of General Psychiatry, 58(3), 231–236. Knitzer, J. (1983). Unclaimed children: The failure of public responsibility to children and adolescents in need of mental health services. Washington, DC: Children’s Defense Fund. Lewandowsky, S., Kalish, M., & Ngang, S. K. (2002). Simpliﬁed learning in complex situations: Knowledge partitioning in function learning. Journal of Experimental Psychology: General, 131, 163–193. Lewinsohn, P. M., Rohde, P., Klein, D. N., & Seeley, J. R. (1999). Natural course of adolescent major depressive disorder, I: Continuity into young adulthood. Journal of the American Academy of Child Adolescent Psychiatry, 38, 56–63. Luckasson, R., & Reeve, A. (2001). Naming, deﬁning, and classifying in mental retardation. Mental Retardation, 39(1), 47–52. Mallory, B. (1995). The role of social policy in life-cycle transition. Exceptional Children, 62, 213–223. Markman, A. B., & Ross, B. H. (2003). Category use and category learning. Psychological Bulletin, 129, 592–613. Osgood, D. W., Foster, E. M., Flanagan, C., & Ruth, G. R. (2005). On your own without a net: The transition to adulthood for vulnerable populations. Chicago and London: The University of Chicago Press. Peterson, B. S., Pine, D. S., Cohen, P., & Brook, J. S. (2001). Prospective, longitudinal study of tic, obsessive-compulsive, and attention-deﬁcit/hyperactivity disorders in an epidemiological sample. Journal of the American Academy of Child and Adolescent Psychiatry, 40(6), 685–695. Pine, D. S., Cohen, P., Gurley, D., Brook, J., & Ma, Y. (1998). The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Archives of General Psychiatry, 55(1), 56–64. Pollack, M. H., Otto, M. W., Rosenbaum, J. F., & Sachs, G. S. (1992). Personality disorders in patients with panic disorder: Association with childhood anxiety disorders, early trauma, comorbidity and chronicity. Comparative Psychiatry, 33, 78–83. Pollack, M. H., Otto, M. W., Rosenbaum, J. F., Sachs, G. S., O’Neil, C., Asher, R., & MeltzerBrody, S. (1990). Longitudinal course of panic disorder: Findings from the Massachusetts General Hospital Naturalistic Study. Journal of Clinical Psychiatry, 51, 12–16. Rao, U., Ryan, N. D., Birmaher, B., Dahl, R. E., Williamson, D. E., Kaufman, J., Rao, R., & Nelson, B. (1995). Unipolar depression in adolescents: Clinical outcome in adulthood. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 566–578.

74

MARYANN DAVIS AND NANCY KOROLOFF

Settersten, R. A., Jr., Frustenberg, F. F., & Rumbaut, R. G. (Eds) (2005). On the frontier of adulthood: Theory, research, and public policy. Chicago: The University of Chicago Press. Stowe, M., & Turnbull, R., III (2001). Tools for analyzing policy ‘‘on the books’’ and policy ‘‘on the streets’’. Journal of Disability Policy Studies, 12, 206–214. Vander Stoep, A., Beresford, S. A., Weiss, N. S., McKnight, B., Cauce, A., & Cohen, P. (2000). Community-based study of the transition to adulthood for adolescents with psychiatric disorder. American Journal of Epidemiology, 152, 352–362. Wagner, M. (1995). Outcomes for youths with serious emotional disturbance in secondary school and early adulthood. The Future of Children: Critical Issues for Children and Youths, 5, 90–112. Wagner, M., Cameto, R., & Newman, L. (2003). A changing population. In: M. Wagner, R. Cameto & L. Newman. Youth with disabilities: A changing population. A report of findings from the National Longitudinal Transition Study (NLTS) and National Longitudinal Transition Study-2 (NLTS2). Menlo Park, CA: SRI International. Available at www.nlts2.org/pdfs/full_report_changepop.pdf Wells, T., Sandefur, G. D., & Hogan, D. P. (2003). What happens after the high school years among young persons with disabilities? Social Forces, 82(2), 803–832.

THE SEARCH FOR COORDINATED, CONTINUOUS COMMUNITYBASED CARE: HOW THE PARALLEL EFFORTS OF THE MEDICAL HOME AND SYSTEMS OF CARE CAN INFORM EACH OTHER David S. Mandell, James P. Guevara and Susmita Pati Between 15 and 18% of children in the United States have a chronic physical, developmental, behavioral, or emotional condition that results in limitations in their functioning and greater use of health services than what typically developing children require (McPherson et al., 1998; Newacheck & Halfon, 1998). Numerous researchers, clinicians, and policy makers have expressed concern that current treatment and service models do not adequately address the needs of these children with special healthcare needs (CSHCN) across the multiple systems in which they require care (American Academy of Pediatrics, 2002a; Dickens, Green, Kohrt, & Pearson, 1992; Center for Mental Health Services, 1999; Sia, 1992). The ﬁnancing, organization, and delivery of children’s healthcare in the United States is complex, and often involves the medical, education, mental health, child welfare, and juvenile justice systems, among others. The poor mechanisms for communication within and among these systems often lead to fragmented, uncoordinated care for the children who need it the most. Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 77–94 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14005-2

77

78

DAVID S. MANDELL ET AL.

There have been at least two separate large-scale responses to this fragmented care. In the Medical Home model, primary care physicians provide comprehensive care and act as coordinators of services across multiple systems (Dickens et al., 1992). The Systems of Care model, developed speciﬁcally for children with serious emotional disorders, is similar to the Medical Home, except that care is managed and coordinated by community mental health professionals (Lourie, Katz-Leavy, & Stroul, 1996; Stroul & Friedman, 1986; Zanglis, Furlong, & Casas, 2000). Literature describing Systems of Care also emphasizes the need to coordinate care across the mental health, education, child welfare, and juvenile justice systems. While both frameworks share the same goals, there has been no mention of the concept of the Medical Home within the Systems of Care literature and vice versa. This lack of interdisciplinary collaboration and evaluation may lead to inefﬁcient efforts to improve care for children with mental health needs, in which approaches found to be effective in one system are ignored in the other. Conversely, unnecessary expense may be incurred if practices found to be ineffective in one system are implemented and evaluated yet again by the other. To date, the Federal Center for Mental Health Services of the Substance Abuse and Mental Health Services Administration has funded 85 System-of-Care Demonstration sites in 46 states and territories, as well as a national evaluation of their impact (Holden, Friedman, & Santiago, 2001). Recent federal funding initiatives (Anonymous, 2004), policy and programmatic changes at the American Academy of Pediatrics (Sia, Tonniges, Osterhus, & Taba, 2004), and efforts to create and test standardized Medical Home models (Cooley & McAllister, 2004; Palfrey et al., 2004) suggest that the ﬁeld is poised to disseminate Medical Home models on a national scale as well. This situation makes it critical that those responsible for the implementation of the Medical Home learn from the dissemination and evaluation experiences of Systems of Care. Conversely, Medical Home research has debated, operationalized, and evaluated the impact of individual components of the Medical Home for different populations using concepts and methods from which Systems of Care research could beneﬁt. The objective of this chapter is to provide an overview of the impact of these two models on health outcomes and quality of care for children with mental health needs and the implications of each body of research for the other model. We provide a brief history of the care of these children in the United States that sets the stage for the need for more encompassing systems for service delivery. We then discuss the development of the System of Care and Medical Home models, summarize evaluation results, and discuss implications and future directions for research.

The Search for Coordinated, Continuous Community-Based Care

79

A BRIEF HISTORY OF CHILDREN’S MENTAL HEALTH TREATMENT IN THE UNITED STATES The belief that children have mental health needs different than those of adults is a relatively recent phenomenon. Systematic ﬁeld studies of mental illness began in the early 19th century (Anthony, Eaton, & Henderson, 1995), although awareness of these illnesses and the recognition of the need for treatment were well established by the 1600s (Grob, 1994). Field studies and census data from mental hospitals in the 1800s reveal few cases of mental illness identiﬁed among children under the age of 16. In the late 19th and early 20th centuries, the recognition that children have speciﬁc mental health needs arose as the result of the conﬂuence of several factors. The Progressive Movement promoted child labor and mandatory public education laws that created legal separations between the role of children and adults (Abbott, 1908; Sutton, 1983). Hall (1905) helped popularize the idea that childhood and adolescence constituted distinct periods of development. Perhaps most pressing, however, was a perceived rise in juvenile delinquency and sexual promiscuity. To address these problems, separate courts were established for juvenile offenders to keep children out of institutions and to provide treatment and rehabilitation (Alper, 1941). Based on the work of Healy and Bronner (1916), researchers and policy makers began to think of juvenile crime as arising from ‘‘mental conﬂicts’’ in children. Institutions such as the Chicago Juvenile Psychopathic Institute and the Boston Psychopathic Hospital were established to care for these children (Horn, 1989). In this context, Child Guidance Clinics (CGCs) were developed as demonstration projects around the United States (Horn, 1989). The CGC comprised a three-person team of a social worker, a psychologist, and a psychiatrist. The social worker assisted schools in identifying ‘‘maladjusted children’’ and conducted home visits; the psychologist conducted assessments of cognitive and emotional functioning; and the psychiatrist provided treatment. The original goal of the CGC was to prevent criminal behavior by reducing emotional maladjustment in juvenile delinquents. Very quickly, focus shifted from delinquency prevention to improving the mental health of schoolchildren. This shift reﬂected the belief that schoolchildren with emotional and behavioral problems were likely to beneﬁt from treatment, whereas treatment for delinquents was considered remedial rather than preventive. The lack of primary care physician involvement in the CGCs can be observed in the early records of the Philadelphia CGC, which recorded discussions regarding service coordination between board members, school

80

DAVID S. MANDELL ET AL.

systems, and psychology and psychiatry programs, but not with primary care physicians (Dreyer, 1976). CGCs remained the mainstay of community mental health treatment for children until 1970, when the National Mental Health Act was amended so that community mental health centers (CMHCs) were required to provide services to children (Grob, 1994). During the 1970s and 1980s, CMHCs subsumed most CGCs (Thompson, 1994). The original goal of CMHCs was to provide secondary prevention, but within a few years of their creation, these centers focused on emotional disturbance and social activism rather than mental illness, based on the theory that mental illness was the result of social inequity (Grob, 1994). A number of medical organizations, including the American Medical Association, expressed concern about the deprofessionalization and demedicalization of mental health treatment (Fink & Weinstein, 1979; Thompson, 1994). Psychiatrists often abandoned their afﬁliations with CMHCs and established private practices or afﬁliations with hospitals (Greenblatt, 1975). As a result of these changes and changes in insurance reimbursement structures, children with serious behavioral and emotional problems were increasingly treated in inpatient settings (Mechanic, McAlpine, & Olfson, 1998; Pottick, McAlpine, & Andelman, 2000), moving away from the community model of treatment that had been the norm.

EVOLUTION OF SYSTEMS OF CARE The origin of the Systems of Care movement can be traced to the growing concern in the 1970s that, due to ﬁnancial incentives for inpatient treatment and the separation of mental health services from other services for children, children with emotional and behavioral problems were increasingly treated in inpatient settings, while the availability of community-based services decreased (Knitzer & Olson, 1982). A System of Care for children’s mental health has been deﬁned as ‘‘a comprehensive spectrum of mental health and other necessary services that are organized into a coordinated network to meet the multiple and changing needs of children and adolescents who are severely emotionally disturbed and their families’’ (Homonoff & Maltz, 1991). The ﬁrst relevant mention of Systems of Care appeared in 1973, and referred primarily to the need for institutional and community systems to coordinate care for adults with serious mental illness (Beigel, Bower, & Levenson, 1973). The 1978 President’s Commission on Mental Health and the publication of Unclaimed Children: The Failure of Public Responsibility

The Search for Coordinated, Continuous Community-Based Care

81

to Children and Adolescents in Need of Mental Health Services (Knitzer & Olson, 1982) resulted in an increased awareness of the failings of the children’s mental health system. Ensuing congressional inquiry culminated in the appropriation of funds for a children’s mental health initiative entitled the Child and Adolescent Service System Program (CASSP), the purpose of which was to reform service delivery (CMHS, 1999). Based on these efforts, a core set of values and principles evolved (Stroul & Friedman, 1986) (Table 1). Federal grants were given to states to improve the infrastructure for implementation. Local CASSP teams, consisting of representatives from mental health and substance abuse, mental retardation, education, and child welfare, were formed to oversee implementation. In 1993, the Center for Mental Health Services of the Substance Abuse and Mental Health Services Administration funded the ﬁrst round of demonstration projects in Systems of Care (CMHS, 1999). These demonstrations involved community mental health professionals coordinating services through regular interdisciplinary team meetings for children across a variety of domains, including mental health, education, juvenile justice, and foster care. Ancillary services to improve access to treatment such as transportation, in-school, and in-home services were also provided. To date, there have been 85 demonstration projects in 46 states and territories (CMHS, 1999).

RESULTS FROM SYSTEMS OF CARE RESEARCH Each Systems of Care demonstration required an evaluation. In general, these evaluations did not include rigorous control groups, with two notable exceptions. The Fort Bragg, North Carolina site, included a 5-year, US$ 80 million evaluation funded by the Department of Defense to determine the systemic, clinical, and functional outcomes of providing a range of individualized and family-centered services. Traditional beneﬁts for dependent children in the Fort Bragg area were replaced with a broad range of services, a single point of entry, comprehensive assessments, and no copayment or beneﬁt limit. The intervention impact was assessed by comparing outcomes at Fort Bragg with those at two other military installations in the southeast. The comparison sites restricted services to outpatient treatment, placement in residential treatment, or treatment in an inpatient hospital setting; regular copayment and beneﬁt limits were in effect at the comparison sites. The results of the evaluation suggested greater coordination of care, more delivery of services, and greater satisfaction among families in the system of

82

DAVID S. MANDELL ET AL.

Table 1.

Principles of Systems of Care and the Medical Home.

Underlying Principle

Family and childcentered care

Coordinated care

Comprehensive care, including early identification and prevention

Child and Adolescent Service System

Medical Home

Families should be full participants in all aspects of the planning and delivery of services Provide individualized services in accordance with the unique need and potential of each child, guided by an individualized service plan Provide services within the least restrictive, most normative, clinically appropriate environment Provide integrated services, with linkage between agencies and programs, and mechanisms for planning, developing, and coordinating services Provide case management to ensure that services are delivered in a coordinated and therapeutic manner, and that children can move through systems in accordance with their changing needs

Provide family-centered care by developing partnerships with families and respecting diversity Share clear, unbiased information with the family about medical care and management, and about specialty and community services

Provide a comprehensive array of services that address the child’s physical, emotional, social, and educational needs Promote early identiﬁcation and intervention for children with emotional problems in order to enhance the likelihood of positive outcomes

Obtain consultations and referrals where appropriate. Establish shared management plans in partnership with the child and family Provide care coordination services in which families, physicians, and other service providers implement a speciﬁc care plan as an organized team Interact with early intervention programs, schools, early education, and child care programs to ensure that needs are addressed Provide primary care including acute and chronic care, preventive services, screenings, and counseling

The Search for Coordinated, Continuous Community-Based Care

83

Table 1. (Continued ) Underlying Principle

Child and Adolescent Service System

Medical Home

Continuity of care

Ensure smooth transitions to the adult service system as children reach maturity

Cultural competency

Provide services without regard to race, religion, national origin, sex, or physical disability. Services should be sensitive and responsive to cultural differences and special needs Protect children’s rights, and promote advocacy efforts for emotionally disturbed children and youth

Provide care over an extended period of time to ensure continuity Plan and organize transitions to other providers and adult care Assure that ambulatory and inpatient care will be continuously available Provide developmentally appropriate and culturally competent health assessments

Advocacy

Data standards

Maintain a complete, accessible, comprehensive, central record that preserves conﬁdentiality

care site, but few differences in clinical outcomes on a variety of measures (Bickman, Lambert, Andrade, & Penaloza, 2000). While a number of researchers questioned these results, citing the lack of generalizability and critiquing the lack of careful measurement of the quantity and quality of treatment and services (Friedman & Burns, 1996; Mordock, 1997; Weisz, Han, & Valeri, 1997), examination of the intervention suggests that it was implemented with ﬁdelity, the outcomes were measured carefully (Bickman, 1996, 1997; Bickman et al., 1998, 2000; Bickman & Noser, 1999), and care was coordinated to a greater extent among children in the demonstration project than among other children (Bickman, Karver, & Schut, 1997; Lambert, Salzer, & Bickman, 1998). Two criticisms of the Fort Bragg study were that the military population limited generalizability of the results and that children were not randomized to the experimental and control groups. To

84

DAVID S. MANDELL ET AL.

address these concerns, a second evaluation took place in Stark County, OH, using a civilian population and a randomized design (Bickman, Noser, & Summerfelt, 1999). In this study, a multi-agency System of Care served participating children within the public mental health system. Participants were randomly assigned to either a group that was immediately eligible to receive services within the System of Care or a group that was required to seek services on its own. The major differences in services were that more children and families in the system of care group received case management and home visits than those in the comparison group. The 18-month outcomes of this study again showed that while access, type, and amount of care were better in the System of Care, there were no differences in changes in symptoms or functioning compared with children who received care outside the site. It is important to note that in neither study was the nature of treatment assessed in any way. A lengthy debate ensued regarding potential explanations for the failure of the system of care model to improve clinical outcomes. Hypotheses have included that (1) systems reform that results in improved coordination and collaboration is too far removed from clinical outcomes to affect them (Salzer & Bickman, 1997); (2) clinicians are not successful in matching children to appropriate services, even when those services are available (Bickman et al., 1997); or (3) mental health treatment for children is not effective as delivered in real world settings, because either clinicians do not implement efﬁcacious treatments or a host of variables that are present in the community but not in experimental settings mediate outcomes (Weisz, Donenberg, Han, & Weiss, 1995). The majority of researchers in this area agree that regardless of the reason for the lack of improved clinical outcomes in Systems of Care sites, future research must collect more sophisticated information about service and treatment processes, and the speciﬁc mechanisms of coordination and collaboration.

EVOLUTION OF THE MEDICAL HOME Sia et al. (2004) recently provided an excellent review of the evolution of the Medical Home from concept to practice. The Medical Home was ﬁrst mentioned in the scientiﬁc literature in 1967 (American Academy of Pediatrics, 1967), but it was not until 10 years later that related policy initiatives evolved. Perhaps the most successful of these initiatives was developed as a result of the pioneering work of Sia and others in the mid-1970s, in response to the growing recognition of the changing nature of morbidity facing

The Search for Coordinated, Continuous Community-Based Care

85

children (Haggerty, Roghmann, & Pless, 1975) and the important role played by the pediatrician in confronting that morbidity (Sia, 1992; Sia & Breakey, 1985). The Medical Home addressed the increasing complexity of children’s needs and the increased fragmentation of pediatric care (Hughes, Grayson, & Stiles, 1977) by advocating for a primary care site that was geographically and ﬁnancially accessible, provided continuous care from the prenatal period through adolescence, identiﬁed additional needs and linked families to appropriate care, and had a community orientation (Sia & Breakey, 1985). In the late 1980s and early 1990s, a number of states began to adopt the Medical Home model and related training (Sia et al., 2004), and by 1992, the American Academy of Pediatrics had formally adopted the Medical Home as ‘‘accessible, continuous, comprehensive, family centered, coordinated, and compassionate’’ care that should be provided for all children (Dickens et al., 1992). By 2002, the AAP had developed a more comprehensive policy statement that outlined 10 key elements of the Medical Home (American Academy of Pediatrics, 2002a). These are presented in Table 1, adjacent to the principles of Systems of Care (Stroul & Friedman, 1986). In the Medical Home, the physician acts as the central ﬁgure and is subsequently involved with all aspects of the child’s care. The physician works closely with the family and appropriate health, school, and community agencies to ensure that children receive optimal care at all times (Dickens et al., 1992). Originally, the Medical Home was applied to children in speciﬁc at-risk groups or with speciﬁc healthcare needs. For example, the ﬁrst Medical Home model was developed for children at risk for child abuse (Sia & Breakey, 1985). Proponents of this model have advocated its importance for many other children with special healthcare needs, including children with health conditions that require care across multiple systems (American Academy of Pediatrics, Committee on Children with Disabilities, 2001a, 2001b), children in foster care (American Academy of Pediatrics, 2002b), and children in special education (Anonymous, 2000). In response to these advocacy efforts, the AAP created the National Center of Medical Home Initiatives for Children with Special Needs in 1999, which organizes training (Moore & Tonniges, 2004) and mentorship efforts (Sia et al., 2004), and the Health Resources and Services Administration (HRSA) has made funds available for states to pilot Medical Home demonstrations (Anonymous, 2004). These efforts have been neither as extensive as the Systems of Care demonstration projects nor required rigorous evaluation of implementation or outcomes.

86

DAVID S. MANDELL ET AL.

RESULTS FROM MEDICAL HOME RESEARCH The processes and outcomes of individual components of the Medical Home, including access to care (Kempe et al., 2000; Shi, Starﬁeld, Politzer, & Regan, 2002; Starﬁeld & Shi, 2004), continuity of care (Christakis, Mell, Koepsell, Zimmerman, & Connell, 2001; Christakis, Mell, Wright, Davis, & Connell, 2000; Christakis, Wright, Koepsell, Emerson, & Connell, 1999; Ortega, Stewart, Dowshen, & Katz, 2000), having a regular source of care (Cunningham & Trude, 2001; Ryan, Riley, Kang, & Starﬁeld, 2001), and coordination of specialty care (Forrest et al., 2000; Gupta, O’Conner, & Quezada-Gomez, 2004), among other components (Starﬁeld, 1998), have been extensively studied using observational research methods, but not more rigorous research designs such as randomized trials. These studies, however, have provided evidence for the development of the critical components of the Medical Home model. One particularly important ﬁnding is that these factors make little difference in quality of care for children in the general population. For example, there are mixed ﬁndings regarding the effects of continuity of care on preventive care such as lead screening and vaccinations (Christakis et al., 1999, 2000; Kempe et al., 2000; Ortega et al., 2000). Differences are more apparent when the population of interest was the one with special healthcare needs (Forrest et al., 2000; Gupta et al., 2004). Fewer studies have examined the availability of all the components of the Medical Home for CSHCN, the ﬁdelity of implementation, and related outcomes. One recent study used the National Survey of CSHCN to provide estimates of the proportion of CSHCN whose care met AAP criteria for a Medical Home, and service outcomes for those whose care met these criteria compared with those whose care did not (Strickland et al., 2004). The results suggest that approximately half of all CSHCN get care that meets criteria for the Medical Home, including having a usual source of care, a personal medical professional, and appropriate referrals for specialty care, coordinated care, and family-centered care. These results are similar to those found using other population-based datasets (Bethell, Read, & Brockwood, 2004). Those whose care met these criteria were half as likely as those whose care did not meet criteria to report delays in care or unmet health needs, and three times more likely to receive family support services when needed. This study did not, however, test the independent effects of different components of the Medical Home on outcomes. Other research has focused on the implementation of Medical Home training. Moore and Tonniges (2004) discuss the development, implementation, and evaluation of a training program to improve physicians’ ability to

The Search for Coordinated, Continuous Community-Based Care

87

implement Medical Home principles in their practices. Pre-post evaluations of participating physicians’ knowledge of Medical Home principles suggested little improvement, although the authors suggest that concurrent Medical Home initiatives may have resulted in a less naı¨ ve sample. Despite the lack of ﬁndings, the authors argue for institutionalization of the training program. At least two groups of researchers have developed speciﬁc models for implementing Medical Home principles in pediatric practices (Cooley & McAllister, 2004; Palfrey et al., 2004). Cooley and McAllister (2004) assessed the feasibility of implementing their Family Centered Medical Home Improvement Model, which addresses six domains: organizational capacity, chronic condition management, care coordination, community outreach, data management, and quality improvement. The authors have implemented this program in many practices across different states, and provided a qualitative assessment of the implementation experience of four of these practices. They identify three critical characteristics of the implementation process, including having a clearly deﬁned role for the care coordinator, involvement of family members, and systematic identiﬁcation of the CSHCN population within each practice. This study did not provide any data on the model’s effect on care processes or clinical outcomes, although the authors acknowledge the importance of studying both. In addition to testing the feasibility of implementation of a different Medical Home model, Palfrey et al. (2004) assessed whether processes improved within practices in which the model was implemented. Their model was developed by the Pediatric Alliance for Coordinated Care, and emphasized family participation, hiring a nurse practitioner who acted as a care coordinator, creating an individualized health plan for each child, streamlining ofﬁce procedures to improve coordination of appointments and ordering of supplies and therapeutics, implementing regular continuing medical education, and expediting referrals and communications with specialists (Silva, Soﬁs, & Palfrey, 2000). The effects of this model on changes in practice and family satisfaction with care were evaluated on 117 CSHCN at six pediatric practices over the course of a 2-year study. Evaluation ﬁndings were generally positive, although mixed, regarding the extent to which improvements occurred as a result of the Medical Home model implementation. More than 60% of parents reported improvements in access to and continuity of a variety of types of care. The proportion responding positively was uniformly higher among parents of severely affected children. Overall satisfaction with care, involvement in decision making, access to written information and basic services, and communication with the physician did not change appreciably pre and post intervention. The authors

88

DAVID S. MANDELL ET AL.

note that satisfaction was very high at baseline, which may account for this lack of change. Children were hospitalized fewer times after the implementation of the program and parents lost fewer days of work. There was no change in the number of emergency room visits or missed days of school among children. These studies provide important information regarding the current state of pediatric practice and the potential for existing Medical Home models to improve that practice. Prior research has been primarily observational or quasi-experimental in nature. There is a strong need to extend this research by studying whether: (1) effective implementation of these components is associated with improved clinical outcomes in a randomized trial and (2) there is beneﬁt associated with having these components delivered from one source, i.e., the Medical Home, rather than from different sources of care (Starﬁeld & Shi, 2004).

IMPLICATIONS OF SYSTEMS OF CARE RESEARCH FOR FUTURE MEDICAL HOME RESEARCH The Medical Home as a framework for pediatric primary care shows promise for improving the quality of care for CSHCN. There is general agreement that pediatricians should provide care that is family and child centered, coordinated, comprehensive, continuous, and culturally competent (American Academy of Pediatrics, 2002a). However, implementing this type of care is not without cost. Care coordination is expensive (Antonelli & Antonelli, 2004), and the addition of other Medical Home components may incur even greater costs. Based on the Systems of Care experience, we may expect the following challenges in implementing the Medical Home. As with Systems of Care, one of the most important concerns for the Medical Home is the careful operationalization of its underlying principles. The principles espoused by both care models include terms such as cultural competency and family-centered care, which are difﬁcult to describe and even more difﬁcult to measure (Cooley & McAllister, 2004; Davidson, Silva, Soﬁs, Ganz, & Palfrey, 2002). To date, there is little research on what speciﬁc actions clinicians should take to implement these principles, and what expected outcomes might result. Various models of the Medical Home have emphasized or omitted different principles (American Academy of Pediatrics, 2002a; Cooley & McAllister, 2004; Palfrey et al., 2004; Starﬁeld & Shi, 2004), and consensus or empirical evidence will be required to determine the relative importance of each principle.

The Search for Coordinated, Continuous Community-Based Care

89

A second area of concern is the careful measurement of processes and outcomes of interest. The majority of research conducted on Systems of Care focused on measuring the system intervention, but neglected to measure the types and quality of treatment (Weisz, 2000; Weisz et al., 1997). It was not possible to determine which components were responsible when interventions did not show desired outcomes. In recognition of this dilemma, the National Institute of Mental Health has created a funding mechanism to study treatment within Systems of Care, and the Center for Mental Health Services has a large-scale effort to implement evidence-based treatment in Systems of Care. As Medical Home research moves forward, it will be important to measure ﬁdelity to Medical Home principles and associated processes (Bethell et al., 2004; Moore & Tonniges, 2004; Strickland et al., 2004), the nature of treatment occurring within practices, and relevant clinical outcomes. A third area of importance is the study of subgroup variation. In general, Systems of Care evaluations did not distinguish between children with different diagnoses in measuring outcomes. Only after the intervention results showed little clinical effect did researchers begin to examine subgroup variation (Diala et al., 2000; Mandell, Walrath, Manteuffel, Sgro, & PintoMartin, 2005; Walrath et al., 2001, 2003, 2004). Investigators were limited, however, by the measures collected from children and their families. It may be that systemic interventions are particularly effective for some groups of children, and efforts should be targeted accordingly. Medical Home research, for example, showed no effect of continuity of care on general pediatric procedures (Ortega et al., 2000). The Medical Home guidelines, however, have focused on its importance for children at risk of abuse, those in foster care, those who have speciﬁc disabling conditions, and those who are medically fragile. Medical Home research should focus on those populations for whom we believe our efforts and providing continuous, coordinated, family-centered care will be particularly effective. There are important differences between primary care and the community mental health service system that may limit the applications of ﬁndings from Systems of Care to Medical Home research. For example, primary care is much more likely to be the ﬁrst point of contact for CSHCN, while the mental health system is usually a much later – or even last – point of contact (Dulcan et al., 1990; Horwitz, Leaf, & Leventhal, 1998). Primary care physicians see a larger and more varied population than mental health professionals, and are responsible for providing a much greater array of services and treatments. These differences may result in differences in the role and importance of collaboration and referral, and the speciﬁcity of related

90

DAVID S. MANDELL ET AL.

system and practice-level interventions. It is also arguable that many treatments provided in primary care have proven more effective than treatments provided in community mental health settings (Weisz, Donenberg, Han, & Kauneckis, 1995; Weisz & Jensen, 1999; Weisz, Weiss, & Donenberg, 1992), although research on conditions such as asthma (Janson & Weiss, 2004) and otitis media (Blomgren & Pitkaranta, 2003) suggests considerable variation in treatment of common pediatric conditions in primary care. If treatments in primary care are more likely to be effective, however, there may be less need to focus on their measurement as part of Medical Home research. Despite these differences, the parallels between these system-level interventions are striking. The experiences and limitations from Systems of Care demonstrations and other controlled studies of behavioral care redesign have important implications for future Medical Home research. Successful implementation of a potentially expensive, system-level Medical Home intervention requires carefully deﬁning key elements and rigorously testing proposed interventions using appropriate experimental designs. Drawing upon past experience with Systems of Care can only enhance our efforts to translate the philosophy of the Medical Home into informative research and, ultimately, into successful practice.

REFERENCES Abbott, E. (1908). A study of the early history of child labor in America. American Journal of Sociology, 14(1), 15–37. Alper, B. (1941). Forty years of the juvenile court. American Sociological Review, 6(2), 230–240. American Academy of Pediatrics. (1967). Pediatric records and a ‘‘medical home’’. In: Standards of child care (pp. 77–79). Evanston, IL: American Academy of Pediatrics. American Academy of Pediatrics. (2002a). The medical home. Pediatrics, 110(1 Pt 1), 184–186. American Academy of Pediatrics. (2002b). Health care of young children in foster care. Pediatrics, 109(3), 536–541. American Academy of Pediatrics, Committee on Children with Disabilities. (2001a). Role of the pediatrician in family-centered early intervention services. Pediatrics, 107(5), 1155–1157. American Academy of Pediatrics, Committee on Children with Disabilities. (2001b). Technical report: The pediatrician’s role in the diagnosis and management of autistic spectrum disorder in children. Pediatrics, 107(5), E85. Anonymous. (2000). Provision of educationally-related services for children and adolescents with chronic diseases and disabling conditions. American Academy of Pediatrics. Committee on Children with Disabilities. Pediatrics, 105(2), 448–451. Anonymous. (2004). Maternal and child health bureau programs. From http://mchb.hrsa.gov:80/ programs/default.htm Anthony, J., Eaton, W., & Henderson, A. (1995). Looking to the future in psychiatric epidemiology. Epidemiologic Reviews, 17(1), 240–242.

The Search for Coordinated, Continuous Community-Based Care

91

Antonelli, R., & Antonelli, D. (2004). Providing a medical home: The cost of care coordination services in a community-based, general pediatric practice. Pediatrics, 113(5), 1522–1528. Beigel, A., Bower, W., & Levenson, A. (1973). A uniﬁed system of care: Blueprint for the future. American Journal of Psychiatry, 130(5), 554–558. Bethell, C., Read, D., & Brockwood, K. (2004). Using existing population-based data sets to measure the American Academy of Pediatrics deﬁnition of medical home for all children and children with special health care needs. Pediatrics, 113(5), 1529–1538. Bickman, L. (1996). Implications of a children’s mental health managed care demonstration evaluation. Journal of Mental Health Administration, 23(1), 107–117. Bickman, L. (1997). Resolving issues raised by the Fort Bragg evaluation. New directions for mental health services research. American Psychologist, 52(5), 562–565. Bickman, L., Karver, M., & Schut, L. (1997). Clinician reliability and accuracy in judging appropriate level of care. Journal of Clinical and Consulting Psychology, 65, 515–520. Bickman, L., Lambert, E., Andrade, A., & Penaloza, R. (2000). The Fort Bragg continuum of care for children and adolescents: Mental health outcomes over 5 years. Journal of Consulting and Clinical Psychology, 68(4), 710–716. Bickman, L., & Noser, K. (1999). Meeting the challenges in the delivery of child and adolescent mental health services in the next millennium: The continuous quality improvement approach. Applied and Preventive Psychology, 8(4), 247–255. Bickman, L., Noser, K., & Summerfelt, W. (1999). Long-term effects of a system of care on children and adolescents. Journal of Behavioral Health Services and Research, 26(2), 185–202. Bickman, L., Salzer, M., Lambert, E., Saunders, R., Summerfelt, W., Heﬂinger, C., & Hamner, K. (1998). Rejoinder to Mordock’s critique of the Fort Bragg evaluation project: The sample is generalizable and the outcomes are clear. Child Psychiatry and Human Development, 29(1), 77–91. Blomgren, K., & Pitkaranta, A. (2003). Is it possible to diagnose acute otitis media accurately in primary health care? Family Practice, 20(5), 524–527. Center for Mental Health (1999). Mental health: A report of the surgeon general. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health. Christakis, D., Mell, L., Koepsell, T., Zimmerman, F., & Connell, F. (2001). Association of lower continuity of care with greater risk of emergency department use and hospitalization in children. Pediatrics, 107(3), 524–529. Christakis, D., Mell, L., Wright, J., Davis, R., & Connell, F. (2000). The association between greater continuity of care and timely measles–mumps–rubella vaccination. American Journal of Public Health, 90(6), 962–965. Christakis, D., Wright, J., Koepsell, T., Emerson, S., & Connell, F. (1999). Is greater continuity of care associated with less emergency department utilization? Pediatrics, 103(4 Pt 1), 738–742. CMHS. (1999). Annual report to congress on the evaluation of the comprehensive community mental health services for children and their families program. Atlanta, GA: ORC Macro. Cooley, W., & McAllister, J. (2004). Building medical homes: Improvement strategies in primary care for children with special health care needs. Pediatrics, 133(5), 1499–1506. Cunningham, P., & Trude, S. (2001). Does managed care enable more low income persons to identify a usual source of care? Implications for access to care. Medical Care, 39(7), 716–726.

92

DAVID S. MANDELL ET AL.

Davidson, E., Silva, T., Soﬁs, L., Ganz, M., & Palfrey, J. (2002). The doctor’s dilemma: Challenges for the primary care physician caring for the child with special health care needs. Ambulatory Pediatrics, 2, 218–223. Diala, C., Muntaner, C., Walrath, C., Nickerson, K., LaVeist, T., & Leaf, P. (2000). Racial differences in attitudes toward professional mental health care and in the use of services. American Journal of Orthopsychiatry, 70(4), 455–464. Dickens, M., Green, J., Kohrt, A., & Pearson, H. (1992). American Academy of Pediatrics ad hoc task force on deﬁnition of the medical home: The medical home. Pediatrics, 90(5), 774. Dreyer, B. (1976). The mental hygiene movement: Institutional response to individual concern. The early years of the Philadelphia child guidance clinic. American Journal of Public Health, 66(1), 85–91. Dulcan, M., Costello, E., Costello, A., Edelbrock, C., Brent, D., & Janiszewski, S. (1990). The pediatrician as gatekeeper to mental healthcare for children: Do parents’ concerns open the gate. Journal of the American Academy of Child and Adolescent Psychiatry, 29(3), 453–458. Fink, P., & Weinstein, S. (1979). Whatever happened to psychiatry? The deprofessionalization of community mental health centers. American Journal of Psychiatry, 136(4A), 406–409. Forrest, C., Glade, G., Baker, A., Bocian, A., von Schrader, S., & Starﬁeld, B. (2000). Coordination of specialty referrals and physician satisfaction with referral care. Archives of Pediatrics and Adolescent Medicine, 154(5), 499–506. Friedman, R., & Burns, B. (1996). The evaluation of the Fort Bragg demonstration project: An alternative interpretation of the ﬁndings. Journal of Mental Health Administration, 23(1), 128–136. Greenblatt, M. (1975). Psychiatry: The battered child of medicine. New England Journal of Medicine, 292(5), 246–250. Grob, G. (1994). The mad among us: A history of the care of America’s mentally ill. Cambridge: Harvard University Press. Gupta, V., O’Conner, K., & Quezada-Gomez, C. (2004). Care coordination services in pediatric practices. Pediatrics, 113(5), 1517–1528. Haggerty, R., Roghmann, K., & Pless, I. (1975). Child health and the community. New York: John Wiley and Sons. Hall, G. (1905). Adolescence; its psychology and its relations to physiology, anthropology, sociology, sex, crime, religion and education. New York: D. Appleton. Healy, W., & Bronner, A. (1916). Youthful offenders: A comparative study of two groups, each of 1,000 young recidivists. American Journal of Sociology, 22(1), 38–52. Holden, E., Friedman, R., & Santiago, R. (2001). Overview of the national evaluation of the comprehensive community mental health services for children and their families program. Journal of Emotional and Behavioral Disorders, 9(1), 4–12. Homonoff, E., & Maltz, P. (1991). Developing and maintaining a coordinated system of community-based services to children. Community Mental Health Journal, 27(5), 347–358. Horn, M. (1989). Before it’s too late: The child guidance movement in the United States, 1922–1945. Philadelphia: Temple University Press. Horwitz, S., Leaf, P., & Leventhal, J. (1998). Identiﬁcation of psychosocial problems in pediatric primary care: Do family attitudes make a difference. Archives of Pediatrics and Adolescent Medicine, 152(4), 367–371. Hughes, J., Grayson, R., & Stiles, F. (1977). Fragmentation of care and the medical home. Pediatrics, 60(4), 559.

The Search for Coordinated, Continuous Community-Based Care

93

Janson, S., & Weiss, K. (2004). A national survey of asthma knowledge and practices among specialists and primary care physicians. Journal of Asthma, 41(3), 343–348. Kempe, A., Beaty, B., Englund, B. P., Roark, R. J., Hester, N., & Steiner, J. F. (2000). Quality of care and use of the medical home in a state-funded capitated primary care plan for low-income children. Pediatrics, 105(5), 1020–1028. Knitzer, J., & Olson, L. (1982). Unclaimed children: The failure of public responsibility to children and adolescents in need of mental health services. Washington, DC: Children’s Defense Fund. Lambert, E., Salzer, M., & Bickman, L. (1998). Clinical outcome, consumer satisfaction, and ad hoc ratings of improvement in children’s mental health. Journal of Consulting and Clinical Psychology, 66(2), 270–279. Lourie, I., Katz-Leavy, J., & Stroul, B. (1996). Individualized services in a system of care. In: B. Stroul (Ed.), Children’s mental health: Creating Systems of Care in a changing society. Baltimore, MD: Brookes. Mandell, D., Walrath, C., Manteuffel, B., Sgro, G., & Pinto-Martin, J. (2005). Characteristics of children with autistic spectrum disorders served in comprehensive community-based mental health settings. Journal of Autism and Developmental Disorders, 35(3), 313–321. McPherson, M., Arango, P., Fox, H., Lauver, C., McManus, M., Newacheck, P., et al. (1998). A new deﬁnition of children with special health care needs. Pediatrics, 102(1 Pt 1), 137–140. Mechanic, D., McAlpine, D., & Olfson, M. (1998). Changing patterns of psychiatric inpatient care in the United States, 1988–1995. Archives of General Psychiatry, 55(9), 785–791. Moore, B., & Tonniges, T. (2004). The ‘‘every child deserves a medical home’’ training program: More than a traditional continuing medical education course. Pediatrics, 113(5), 1479–1484. Mordock, J. (1997). The Fort Bragg continuum of care demonstration project: The population served was unique and the outcomes are questionable. Child Psychiatry and Human Development, 27(4), 241–254. Newacheck, P., & Halfon, N. (1998). Prevalence and impact of disabling chronic conditions in childhood. American Journal of Public Health, 88(4), 610–617. Ortega, A. N., Stewart, D. C., Dowshen, S. A., & Katz, S. H. (2000). The impact of a pediatric medical home on immunization coverage. Clinical Pediatrics, 39(2), 89–96. Palfrey, J., Soﬁs, L., Davidson, E., Liu, J., Freeman, L., & Ganz, M. (2004). The pediatric alliance for coordinated care: Evaluation of a medical home model. Pediatrics, 113(5), 1507–1516. Pottick, K., McAlpine, D., & Andelman, R. (2000). Changing patterns of psychiatric inpatient care for children and adolescents in general hospitals, 1988–1995. American Journal of Psychiatry, 157(8), 1267–1273. Ryan, S., Riley, A., Kang, M., & Starﬁeld, B. (2001). The effects of regular source of care and health need on medical care use among rural adolescents. Archives of Pediatrics and Adolescent Medicine, 155(2), 184–190. Salzer, M., & Bickman, L. (1997). Delivering effective children’s services in the community: Reconsidering the beneﬁts of system interventions. Applied and Preventive Psychology, 6, 1–13. Shi, L., Starﬁeld, B., Politzer, R., & Regan, J. (2002). Primary care, self-rated health, and reductions in social disparities in health. Health Services Research, 37(3), 529–550. Sia, C. (1992). Abraham Jacobi Award address, April 14, 1992. The medical home: Pediatric practice and child advocacy in the 1990s. Pediatrics, 90(3), 419–423. Sia, C., & Breakey, G. (1985). The role of the medical home in child abuse prevention and positive child development. Hawaii Medical Journal, 44(7), 242–244, 247.

94

DAVID S. MANDELL ET AL.

Sia, C., Tonniges, T., Osterhus, E., & Taba, S. (2004). History of the medical home concept. Pediatrics, 113(5), 1473–1478. Silva, T., Soﬁs, L., & Palfrey, J. (2000). Practicing comprehensive care: A physician’s manual for implementing a medical home for children with special health care needs. Boston, MA: Institute for Community Inclusion, Children’s Hospital. Starﬁeld, B. (1998). Primary care: Balancing health needs, services, and technology. New York, NY: Oxford Press. Starﬁeld, B., & Shi, L. (2004). The medical home, access to care, and insurance: A review of evidence. Pediatrics, 113(5), 1493–1498. Strickland, B., McPherson, M., Weissman, G., van Dyck, P., Huang, Z., & Newacheck, P. (2004). Access to the medical home: Results from the national survey of children with special health care needs. Pediatrics, 113(5), 1485–1492. Stroul, B., & Friedman, R. (1986). A system of care for severely emotionally disturbed children and youth. Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Sutton, J. (1983). Social structure, institutions, and the legal status of children in the United States. American Journal of Sociology, 88(5), 915–947. Thompson, J. (1994). Trends in the development of psychiatric services, 1844–1994. Hospital and Community Psychiatry, 45(10), 987–992. Walrath, C., Petras, H., Mandell, D., Stephens, R., Holden, E., & Leaf, P. (2004). Gender differences in patterns of risk factors among children receiving mental health services: Latent class analyses. Journal of Behavioral Health Services and Research, 31(3), 297–311. Walrath, C., Ybarra, M., Holden, E., Liao, Q., Santiago, R., & Leaf, P. (2003). Children with reported histories of sexual abuse: Utilizing multiple perspectives to understand clinical and psychosocial proﬁles. Child Abuse and Neglect, 27, 509–524. Walrath, C. M., Mandell, D. S., Liao, Q. H., Holden, E. W., De Carolis, G., Santiago, R. L., et al. (2001). Suicide attempts in the ‘‘comprehensive community mental health services for children and their families’’ program. Journal of the American Academy of Child and Adolescent Psychiatry, 40(10), 1197–1205. Weisz, J. (2000). Agenda for child and adolescent psychotherapy research: On the need to put science into practice. Archives of General Psychiatry, 57(9), 837–838. Weisz, J., Donenberg, G., Han, S., & Kauneckis, D. (1995). Child and adolescent psychotherapy outcomes in experiments versus clinics: Why the disparity? Journal of Abnormal Child Psychology, 23(1), 83–106. Weisz, J., Donenberg, G., Han, S., & Weiss, B. (1995). Bridging the gap between laboratory and clinic in child and adolescent psychotherapy. Journal of Consulting and Clinical Psychology, 63(5), 688–701. Weisz, J., Han, S., & Valeri, S. (1997). More of what? Issues raised by the Fort Bragg study. American Psychologist, 52(5), 541–545. Weisz, J., & Jensen, P. (1999). Efﬁcacy and effectiveness of child and adolescent psychotherapy and pharmacotherapy. Mental Health Services Research, 1(3), 125–157. Weisz, J., Weiss, B., & Donenberg, G. (1992). The lab versus the clinic. Effects of child and adolescent psychotherapy. American Psychologist, 47(12), 1578–1585. Zanglis, I., Furlong, M., & Casas, J. (2000). Case study of a community mental health collaborative: Impact on identiﬁcation of youths with emotional or behavioral disorders. Behavioral Disorders, 25(4), 359–371.

EVALUATING SERVICE SYSTEM COORDINATION FROM THE PROVIDERS’ PERSPECTIVE Denine Northrup The current philosophy of behavioral health service systems touts the necessity of coordination of various programs and supports to adequately meet the multifaceted needs of consumers and their families to produce optimal outcomes. In youth behavioral health services, the system-of-care philosophy has been heralded as critical in addressing many service delivery challenges. The Child and Adolescent Service System Program (CASSP) initiated by the National Institute of Mental Health to support state and local systems of care for children with severe emotional disturbance (SED) and their families (Stroul & Friedman, 1986). CASSP principles have become widely accepted in the area of service delivery promoting services that are child-centered, family focused, and community-based. For adults, the recovery model has permeated systems of care through promotion in the Presidential New Freedom Commission’s Report as well as new standards in JCAHO’s 2006–2007 Comprehensive Accreditation Manual for Behavioral Health Care (Joint Commission on Accreditation of Healthcare Organizations, 2006). Recovery model tenets highlight care coordination that is person-centered, culturally competent, and strength-based and that instills a sense of hope and autonomy for consumers (Anthony, Cohen, & Farkas, 2005; Anthony, 2005; Stroul, 1989). Despite the widespread acceptance of the values espoused by these philosophies, the implementation of these values in systems of care is much Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 95–116 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14006-4

95

96

DENINE NORTHRUP

less clear. Behavioral health service systems typically remain fragmented for a variety of reasons related to policy, planning, established infrastructure, and funding. The result is limited service coordination and continuity of care for consumers and their families. As many initiatives develop around the country, it is the role of health services researchers and evaluators to monitor progress and document effective strategies and initiatives as well as to identify initiatives that are not effective so as to promote improvement in behavioral health systems. This chapter will describe one effort to evaluate system change with the initiation of a system of care demonstration project. Recently, similar efforts have assessed services integration in the community for vulnerable populations in both the public and private sectors (Provan, Isett, & Milward, 2004; Provan & Milward, 2001, 1995). The larger evaluation incorporated multiple methods with multiple sources of information, however, this chapter will emphasize the interorganizational evaluation conducted based on provider perspectives. This evaluation strategy was an attempt to measure the potential change in system organization and coordination as the system tried to embody system of care principles. The participatory evaluation incorporated both formative and summative components. The formative evaluation products were available to the evolving system stakeholders and provided critical information about the perceptions of providers across the system. The potential for the use of this information in planning and decision-making hold much hope for system improvement and ongoing enhancement. Ultimately, however, the system’s impact will need to be assessed via client outcomes to determine whether system change affected client-level outcomes. The evaluation presented here focused on a system of care demonstration project that occurred in a large city in the southeastern United States. The mission of the demonstration was to implement a coordinated child-centered and family-focused system-of-care to enable children with SED to be cared for in their homes, schools, and communities, and for children and families to develop skills for managing their lives in their homes and communities.1

INTERORGANIZATIONAL NETWORK APPROACH Network analysis involves the examination of the structure and patterning of relationships among a set of persons or agencies by taking into account the relations or ties that exist, as well as those that do not, among various groups (Knoke & Kuklinski, 1982). For the evaluation project, the structure of relationships within the county behavioral health service network, were

Evaluating Service System Coordination from the Providers’ Perspective

97

examined over the course of three years. The method applied interorganizational theory and a network analytic approach, based on the works of Heﬂinger and Northrup (Heﬂinger, 1996; Heﬂinger & Northrup, 1998, 2000a), Morrissey and colleagues (Morrissey et al., 1994, 1998; Morrissey, 1992; Morrissey, Tausig, & Lindsey, 1985), Van de Ven (1976), and Bolland and Wilson (1991a, 1991b, 1994). Interviews with providers and agencies from the county’s behavioral health service network were conducted at two points in time and gathered providers’ perspectives on the coordination of behavioral health services for children and adolescents. The comparison of perspective prior to and after the implementation of the demonstration system-of-care assesses changes in both the structure and the effectiveness of the system of care. This chapter is using the network analysis data as an illustration of the type of information that can be used both to plan and improve a system or care as well as to evaluate changes in a system’s structure and effectiveness of the system. This network analysis study used information about patterning of ties/ relationships among all agencies in the behavioral health network in the county to determine positions or roles within the system and to describe the relations among these positions or subgroups within the network (Knoke & Kuklinski, 1982).2 Because complex problems, such as addressing the behavioral health needs of children and their families, are sometimes beyond the scope of any one organization, agency, or individual providers often join hands to accomplish mutual goals. When organizations join or cluster together and (a) behave in such a way as to attain a collective goal; (b) develop interdependent processes or specialized roles among members; and/or (c) when perceived as a whole, act as a unit with its own identity, then the formation of these organizations is called a network (Heﬂinger & Northrup, 1998, 2000b). To reach the desired end, the network system takes on structure and process for organizing the activities of its members. Structure refers to administrative procedures that deﬁne roles among members. Process deals with the ﬂow of activities and denotes both the direction and frequency of resource and information exchange between members (Heﬂinger, 1996). Both of these issues are critical when studying the implementation of a system of care. In analytic terms, structural dimensions of an interorganizational network are density, centrality, cluster membership, and fragmentation (Bolland & Wilson, 1994; Van de Ven & Ferry, 1980). These dimensions characterize coordination and interagency dependency. Density describes the proportion of relationships present among network members. Centrality of the network refers to the overall positions held within the network by speciﬁc actors. Cluster membership relates to the patterns of relationships between network

98

DENINE NORTHRUP

members. Fragmentation refers to the cliquishness of a network, the extent to which relationships occur only within clusters instead of between members of different clusters. ‘‘Process,’’ an inextricable part of structure, has as two major identifying characteristics, resource and information ﬂows. Resource ﬂows can be money, physical space, clientele, or materials that are transferred between organizations. Verbal or written communications are also examples of information ﬂows. All of these domains emphasize the structure and process of a network. Ultimately, the effectiveness of the network is of signiﬁcant concern. Dimensions of effectiveness include accessibility, adequacy, quality, and effectiveness of services. Critical to the measurement of these dimensions are the perceptions of each agency involved as to how well other agencies carried out their commitments and whether the relationships were judged to be worthwhile, productive, and satisfying. While it is not being tested in this chapter, it has been hypothesized that the effectiveness of an interagency network can be predicted from variables of network structure.

‘‘BOUNDING’’ THE NETWORK AND CHOOSING SURVEY RESPONDENTS Applying this interorganizational network framework to a ‘‘real’’ network involves (a) identifying the pertinent organizations that make up the network, (b) identifying the boundary spanners or key informants within the system, and (c) collecting data from questionnaires that tap both structural and effectiveness dimensions. Before any service network can be evaluated, the behavioral health network must be deﬁned. Agencies and individual service providers must be identiﬁed, and the network then must be ‘bounded’ to determine which agencies and service providers will be included in the description of system structure and to determine who will serve as survey respondents. The behavioral health network in the demonstration for youth and their families included behavioral health service agencies within the county. However, the behavioral health ‘‘system’’ also includes other actors. Each behavioral health service organization is afﬁliated, whether formally or informally, with a variety of other community agencies in order to secure or send client referrals, jointly offer programs, give or receive funding or payment for services, train and recruit staff, supply equipment and other material goods, lobby and perform advocacy work, and engage in other endeavors. The behavioral health system for children and youth is directly linked to other service sectors such as education, juvenile justice, social

Evaluating Service System Coordination from the Providers’ Perspective

99

services, and health. It is important for these other sectors and the corresponding community agencies to describe and rate the effectiveness of the behavioral health network for youth in their community. Identifying the appropriate person within these community agencies to act as survey respondent was the next step in effectively conducting the network analysis. To expedite this process, each agency was asked to nominate the staff person most knowledgeable about their agency’s working relationships within the behavioral health network. In addition, the person ﬁlling out questionnaires was allowed to consult with others to provide the response representative of their agency. At baseline (Spring, 2000), a comprehensive list was compiled of all community agencies and private providers that provided or interacted with behavioral health services for children and youth and their families in the community. Community providers were very receptive to participating in this process as a way to provide feedback about the current system. Followup data collection occurred two years after implementation of the system of care demonstration between November, 2002 and April, 2003. Questionnaires Participants completed two questionnaires and participated in a brief interview. The interview covered the speciﬁcs of the services offered by the agency, the funding sources for the services, the clients served, and provider perspectives on the strengths and challenges that faced the current behavioral health system for children and families. The Network Study Questionnaire (NSQ) examined the structure and pattern of relationships between the various agencies in the network of services for youth with SED and their families in the county. The NSQ is a selfadministered instrument measuring the linkages between the respondent’s agency and each of the organizations in the behavioral health network. For each of the six questions described below, the agency respondent answered for all the other agencies in the network. Staff Interaction (Question 1): Respondents rated the frequency of interaction between their staff and the staff of each other agency on a scale from 0 (not at all) to 4 (very often). Referrals: Respondents indicated whether or not they referred clients to each of the other agencies (Question 2), and whether or not they received client referrals from each of the other agencies (Question 3), on a scale from 0 (not at all) to 4 (very often).

100

DENINE NORTHRUP

Activity Coordination (Question 4): Respondents rated the extent to which the activities of their own program or agency were coordinated with those of other community agencies on a scale from 0 (not at all) to 4 (very well coordinated). Formal Agreement (Question 5): Respondents reported (yes or no) whether or not they had a formal agreement (through contract, memo of understanding, or other legal mandate) with each of the other agencies. Critical Agencies (Question 6): Respondents reported which other agencies or providers it would be critical for their agency to work with in order to promote an ideal service system, on a scale from 0 (not critical) to 4 (very critical).

0 Not at all 1 2 3 4 Very often For two agencies, their responses could reflect no relationship (0), a weak relationship (1 2) or a strong relationship (3 4).

The Assessing Local Service Systems (ALSS) questionnaire was designed to measure the effectiveness of the network by assessing the accessibility, adequacy, quality, and effectiveness of services at a particular point in time, and also assesses the performance of and changes in a system (Morrissey, 1992). Items were grouped together into sections3 that addressed:

quality of services delivered to children and families, service delivery problems encountered by consumers, system performance, and extent to which the system meets the goals of an ideal system. Data Collection Procedure

At baseline, all the providers in the bounded network were mailed information about the study and a self-addressed stamped postcard to respond with their willingness to participate and to indicate best times to be reached. Appointments were scheduled for an evaluation staff member to go to the

Evaluating Service System Coordination from the Providers’ Perspective

101

provider’s agency to conduct the interview and remain available while the provider completed the ALSS and NSQ. Ninety-two service providers and agencies were identiﬁed to be potential participants, of these, 90 participated in providing feedback, resulting in a response rate of 98%. In order to streamline data collection for participants at follow-up, all individual providers were omitted from the network analysis questionnaire, leaving those groups and agencies that serve youth with SED and their families as the core group of participants. At follow-up, 72 provider agencies were identiﬁed, including one new agency not in existence two years prior. At follow-up, both the NSQ and the ALSS surveys were mailed to the respondents to be completed independently. The project manager then contacted each agency by phone to insure that both surveys were received and to answer any questions. Surveys were either returned by mail or picked up by the site coordinator. Of the 72 agencies identiﬁed, 59 agencies responded to the follow-up Nashville Connection Network Analysis Questionnaire resulting in a response rate of 82%. Three providers who did not participate but who were central actors in the network at baseline were included in the follow-up network by using a data imputation strategy.4 Participants The participants were service providers and representatives of agencies from a variety of sectors serving youth between the ages of 8 and 13 with SED. The participants were primarily afﬁliated with private not for proﬁt (57%) and public (22%) agencies. Their principal job responsibilities were supervision (58%) and administration (49%). Seventy-three percent had masters or doctoral level training, and 62% held licensure or certiﬁcation in their ﬁelds. The service sectors represented by the participating agencies are depicted in the graphic below (Fig. 1).

DATA ANALYSIS AND RESULTS ASSESSING EFFECTIVENESS OF NETWORK The ALSS assesses the effectiveness of the network by examining quality, barriers to care, system performance, and how well the system meets the goals of a system of care. In the analyses that compare responses from baseline to follow-up, the sample was limited to the 57 agencies that participated at both points. Providers rated an overall assessment for each of four domains. Fig. 2 compares the baseline and follow-up responses for each of the four domains. Ratings in two of the four domains (System

102

DENINE NORTHRUP Other 14%

Faith-based 1% Juvenile Justice 5%

Mental Health 38%

Health 7% Education 10%

Social Svcs 25%

Fig. 1.

Distribution of Sectors Serving Youth.

5.0

Poor (1) -- Very Good (5)

4.5 Baseline

4.0

Follow-up 3.5 3.0

Significant Improvement

2.48 2.55

2.5

2.22

2.43

Significant Improvement

2.46

2.65

1.88 1.82

2.0 1.5 1.0 Quality

Fig. 2.

Barriers

System Performance

Goals

Overall Ratings Across Domains.

Performance and Goals of an Ideal Service System) reﬂected signiﬁcant improvement from baseline to follow-up as perceived by providers. In addition, each domain was examined in detail to assess the highest and lowest rated areas. At baseline this information can be used to identify areas of attention and devote resources to improve those areas. Recent research has similar information to promote improvement in service capacity for a

Evaluating Service System Coordination from the Providers’ Perspective

103

network of services for chronic disease (Provan, Veazie, Teufel-Shone, & Huddleston, 2004). With the follow-up data, stakeholders can determine whether they have made improvements where they intended and identify new or renewed efforts based on follow-up data. For example, the system performance domain included 39 performance areas. Stakeholders can clearly identify the areas providers perceive as being strengths and those areas that are challenging (see Table 1). One of the strengths of this measurement approach is that there is reasonable variability in ratings across items without any ceiling or ﬂoor effects. For this reason, change may be more readily detected. As is evident in the item level ratings, there is room for improvement even in areas rated more highly. Finally, the areas that have improved can also be documented. This is critical information for system planners and opportune in a participatory evaluation. Highest Rated Domains (Adequate) Complete/thoroughly written treatment plans Grievance mechanisms Children and families feel welcome Active involvement of parents in services Strengths-based approach Lowest Rated Domains (Fairly Poor): Transition to adult services Services for those without insurance Long range funding No long waitlists or scheduling delays Preventing multiple service children from falling through the cracks Domains with Improved Performance from Baseline to Follow-up Preventing children from falling through the cracks Common tracking system Up to date resource directory Joint planning between agencies/providers Preventing state custody to obtain services Timely access to records Coordinated services Ongoing evaluation of services Individualized services Parents take leadership role in case planning Youth in most appropriate setting Individualized, strengths based services

104

Table 1.

Service System Performance.

How Well Does the Service System Perform in the Following Areas?a Service System Performance

Follow-upb

Mean

SD

Mean

SD

2.06 2.12 2.42 2.00 1.98 2.00 2.10 2.50 2.15 2.59 3.11 2.71 2.52 1.72 1.77 1.98 1.88 1.79

0.68 0.76 0.86 0.77 0.63 0.70 0.86 0.86 0.76 1.09 0.92 0.90 0.82 0.60 0.78 0.75 0.68 0.83

2.26 2.09 2.49 2.22 1.89 2.06 2.20 2.54 2.33 2.67 3.08 2.88 2.83 2.00 2.02 2.28 2.20 2.25

0.74 0.77 1.02 0.66 0.70 0.89 0.83 0.86 0.83 0.99 1.02 0.99 0.98 0.71 0.75 0.77 1.00 0.81

Statistically Signiﬁcant Improvement

a a a a a

DENINE NORTHRUP

Expanding services to meet growing needs Appropriate services available to all in need High quality MH treatment services Supportive services available No long waiting lists or scheduling delays Minimizing enrollment red tape Providing transportation Easily accessible service locations Evening and weekend hours Reasonable cost Children and families feel welcomec Priority to services for children and families Community-based treatment options Multiple service children do not fall through the cracks Common tracking system Up to date resource directory Joint planning between agencies/providers Preventing state custody to obtain services

Baselineb

a

2.00 2.11 2.00 2.06 2.25 2.02 1.70 2.15 2.80 2.37 2.36 2.83 2.38 3.19 2.79 2.05 2.10 2.31 2.42 1.83 1.76 2.22

Rated from 1 (very poorly) to 5 (very well). Ratings based on agencies who responded at both baseline and follow-up. c At least 20% of respondents responded ‘‘don’t know’’. b

0.75 0.84 0.83 0.77 0.90 0.71 0.65 0.85 0.88 0.82 0.93 0.92 0.95 0.89 0.93 0.82 0.77 0.91 0.75 0.80 0.77 0.53

2.20 2.57 2.51 2.24 2.62 2.31 1.85 2.60 3.23 2.67 2.70 3.09 2.77 3.38 2.94 2.47 2.43 2.49 2.69 1.86 1.85 2.43

0.76 0.78 0.94 0.85 0.87 0.86 0.74 0.92 0.97 0.92 1.07 0.87 1.03 0.87 0.89 0.98 0.80 0.80 0.80 1.03 0.99 0.72

a

a

a

a

a

a

a

a

Evaluating Service System Coordination from the Providers’ Perspective

Fostering understanding of whole system Timely access to records Minimizing conﬂicting rules/requirementsc Interagency agreements Meaningful discharge between inpatient and community MHCc Coordinated services Long range funding On-going evaluation of services/system Grievance mechanismsc Individualized services Difﬁcult clients adequately servedc Active involvement of parents in services Parents take leadership role in case planning Complete/thorough written treatment plans Strength-based approach Integrated services for substance abuse and mental healthc Youth in most appropriate setting Shared responsibilities among providers Individualized, strength-based services Services for those without insurance beneﬁtsc Transition to adult servicesc Overall performance

105

106

DENINE NORTHRUP

DATA ANALYSIS AND RESULTS ASSESSING STRUCTURE AND PROCESS OF NETWORK While the ALSS generated some useful information on the effectiveness of the system, the NSQ generates a very different type of information. Not only does it directly target the coordination of the network but also uses dyadic relationship data. In other words, not only does it learn from Provider A that they send referrals to Provider B, but also whether Provider B reports receiving referrals from Provider A. While many of the analyses are descriptive in nature, using dyadic data presents a complementary view of the behavioral health system with built in reliability checks. The ﬁrst step in understanding the relationships among agencies in the behavioral health network was to examine the organization and structure of the service delivery system. This was accomplished through a formal network analysis using data from the NSQ. The analyses examined the behavioral health service system network as it existed in 2000, prior to the implementation of the demonstration system of care as well as the network in 2003. Changes in the networks over time were also addressed. By examining the organization and structure of services, several facets of service coordination can be studied. First, the overall number of interagency relationships can be examined to assess the opportunities available for service coordination. The importance of individual agencies and the nature of relationships between agencies also can be determined. As there is no single focal agency in network analysis, information on every pair of relationships is gathered and used to develop an overall picture of the network.5 Measures of Structure and Coordination Several traditional measures of network structure were calculated at each time point. Density, agency centrality, cluster membership, and the coordination index have been identiﬁed as measures that are helpful in describing a network’s overall structure and the extent to which it corresponds to an ‘‘ideal’’ coordinated system of care (Bolland & Wilson, 1994). These measures also provide a means of examining change over time with regard to the patterns of agency functions in interagency coordination of services for children and adolescents and their families in this community. Density The density of a matrix refers to the extent of linkage among the agencies and service providers in the service system, measured as a proportion or

Evaluating Service System Coordination from the Providers’ Perspective

107

ratio of the number of actual links among agencies over the total possible links among agencies (Knoke & Kuklinski, 1982). In other words, the density of a network is a measure of the proportional number of the relationships between the many service providers and agencies in that community network. The greater the number of linkages or reported relationships between agencies, the higher the density. According to Bolland and Wilson (1994), the importance of density as an indicator of coordination can be noted by the simple assumption that ‘‘more relationships reﬂect better coordination,’’ or at least the opportunity for better coordination. Density reﬂects the relative presence or absence of relationships between agencies and ranges from 0 (absence of any links) to 1 (presence of all possible links). The proportion is comparable to a percentage: If the density index is .36, that indicates that in 36% of all possible relationships between agencies, relationships were reported as present. In addition, it indicates that in 64% of the possible relationships between agencies, relationships were reported as absent. Density

3 relationships present

= 1.0 or 100%

3 possible 1 relationships present = .33 or 33% 3 possible

As shown in Table 2, the proportion of strong relationships present ranges between 11 and 26%. While there is an increase in the densities Table 2.

Network Density at Baseline and Follow-upa. 2000 (n ¼ 90)

Staff interaction Conﬁrmed referrals Activity coordination Ideal service system a

0.20 0.11 0.12 0.25

(0.40) (0.22) (0.32) (0.43)

2003 (n ¼ 62) 0.26 0.16 0.18 0.35

(0.44) (0.26) (0.38) (0.48)

For the all networks, density was calculated to reﬂect only the strongest relationships for each matrix by dichotomizing the valued data (originally reported on a scale from 0 (no relationship) to 4 (very strong relationship)) with ratings greater than 2 representing a strong relationship. Density is affected by the number of total possible links, thus, if the number of actors in a network changes over time, density may be directly impacted.

108

DENINE NORTHRUP

Table 3.

Average Strength of the Relationship. Mean Agency Ratinga

Staff interaction Send referrals Receive referrals Activity coordination Ideal service system

2000 (n ¼ 90)

2003 (n ¼ 62)

1.15 0.81 0.62 0.69 1.32

1.48 1.15 0.78 0.99 1.73

a

Agencies are rated from 0 (no relationship with other agency) to 4 (strong relationship with other agency) by all other agencies.

between baseline and follow-up, this could be directly attributable to the reduction in the network size. In addition, the question about the ideal service system does not reﬂect the current status of the network, but rather what the ideal network relationships might look like. Given this, it should not be assumed that an optimal network is one in which every possible relationship exists (density equal to 1.0). Rather, in the ideal world according to the participating providers, the densities for 2003 should be closer to .35, and the densities from 2000 should be closer to .25. If the gap between actual and ideal is diminishing, then the network is becoming more coordinated. In this system of care, a gap remains between actual interaction and ideal levels of interaction as perceived by providers. An additional measure of network activity related to density was also calculated for each network. The level of network activity indicates the strength of the relationship as rated by all respondents about each particular agency or provider. The level of staff interaction is the mean rating of the strength of the relationships across all agencies and demonstrates an increase from the 2000 rating of 1.15–1.48 on a ﬁve-point scale from 0 (no staff interaction between agencies) to 4 (very frequent interaction; see Table 3). Note the difference between all mean ratings as compared with the ideal service system, which suggests that much more intensive interaction and activity between providers would be desirable. While staff interaction approaches the ideal, activity coordination and referrals lag behind. Agency Centrality Measures of agency centrality in a network describe the extent to which the agency is ‘‘important.’’ Important agencies are located in strategic locations

Evaluating Service System Coordination from the Providers’ Perspective

Table 4.

109

Centrality: Betweenness for Staff Interaction.

Juvenile Court Metro Public Schools County Children’s Services Metro Social Services Community Services Agency Community Mental Health Center A Community Mental Health Center B Community Mental Health Outpatient Community Mental Health School Based Co-occurring Services Adolescent Shelter and Outreach Services Advocacy Organization Private Mental Health Services A Sexual Abuse/Domestic Violence Services Private Mental Health Services B

Staff Interaction 2003

2003 Rank

Staff Interaction 2000

2000 Rank

1.000 0.860 0.737 0.533 0.525 0.299

1 2 3 4 5 6

0.679 1.000

6 1

0.787 0.707

3 4

0.244

7

0.618

9

0.239

8

0.238

9

0.233

10 0.859

2

0.680 0.642 0.630

5 7 8

0.600

10

in the network. Agency centrality can be viewed as an index of power in an interorganizational network (Mizruchi & Galaskiewicz, 1993). Organizations with high network centrality are more likely to be in an inﬂuential position and can affect the local service system and control the ﬂow of resources (Oliver & Montgomery, 1996). An agency that is centrally located within a network would be, by virtue of its position, well coordinated with regard to information and resource ﬂows. The centrality coefﬁcients (see Tables 4 and 5) varied from 0 (low centralization) to 1 (high centralization). Two speciﬁc measures of centrality were assessed: ‘‘Betweenness’’ (Freeman) centrality and ‘‘Eigenvector’’ (Bonacich) centrality. Both measures address different types of inﬂuential roles. To illustrate use of these measures, results from the staff interaction network are presented. Betweenness (Freeman) is the number of times an agency serves as an intermediary between two other agencies – broker or liaison role. Agencies in this role are

110

DENINE NORTHRUP

Table 5. Centrality: Eigenvalue for Staff Interaction.

Juvenile Court County Children’s Services Community Services Agency Advocacy Organization Community Mental Health School Based Metro Social Services Community Mental Health Outpatient Community Mental Health Services B State Inpatient Facility State Residential Facility Metro Public Schools Private Mental Health Services B Sexual Abuse/Domestic Violence Services Private Mental Health Services C Private Mental Health Services A

Staff Interaction 2003

2003 Rank

Staff Interaction 2000

2000 Rank

1.000 0.960 0.825 0.766 0.746

1 2 3 4 5

0.980 1.000

2 1

0.881 0.751

4 10

0.737 0.734

6 7

0.871 0.781

5 8

0.714

8

0.708 0.696

9 10

0.766 0.910 0.811 0.801

9 3 6 7

0.751 0.751

10 10

important in terms of information sharing to aide in communication throughout a network.

Betweenness (Freeman)

The most central agency in staff interaction activities at follow-up was the County Juvenile Court. At baseline, this position had been held by the public schools (see Table 4). The top ten agencies at baseline and follow-up that would be critical in terms of information dissemination are shown in Table 4. As these data indicates, traditional mental health agencies were not the only agencies that are necessary to disseminate information throughout the system; rather related service sector agencies, such as the courts and schools, could play a major role in an effective information dissemination effort.

Evaluating Service System Coordination from the Providers’ Perspective

111

Bonacich (Eigenvector)

Bonacich (eigenvector) centrality describes an agency’s centrality to the extent that it is connected to other agencies that are central. Agencies in this role are powerful, since they are connected to other ‘‘well-connected’’ agencies. As shown in Table 5, the Juvenile Court and County Children’s Services held the two most central roles. The agencies who are central actors in network change over time. A number of the highly ranked agencies at baseline were no longer considered as central actors at follow-up, indicating either omission from the network, reduced participation, or providers not perceiving the agencies as central. Cluster Membership and the Coordination Index Coordination is directly linked to the ability of all agencies to interact with other agencies in a network. Coordination of services can only occur if agencies are interacting with each other. The measures of network structure described above focus on the global network that includes all members of the behavioral health service delivery system, however, the patterns in which subsets of agencies work more closely together is also of interest. Through cluster analysis, a network can be divided into subnetworks or clusters of agencies that report the strongest interrelationships. The number of members in clusters and the identity of the cluster members yielded important information about how the network ‘‘does business.’’ The interpretative task of cluster analysis is performed at two levels: (1) within-group in terms of the attributes of the clustered agencies and (2) between-group in terms of relationships between clusters. In the evaluation process, identiﬁcation of the clusters for stakeholders can be very helpful. It quickly identiﬁes the agencies that are on the periphery of the network – some of whom may be critical. In addition, it identiﬁes the groups who are just interacting within their cluster. The optimal network will optimize both within cluster relationships and between cluster relationships. Just as densities (proportion of connections) are

112

DENINE NORTHRUP

Table 6.

Cluster Cluster Cluster Cluster

I II III IV

Staff Interaction Within and Between Cluster Densities. Cluster I

Cluster II

Cluster III

Cluster IV

0.39 0.15 0.24 0.07

0.15 0.44 0.25 0.06

0.24 0.25 0.35 0.18

0.07 0.06 0.18 0.50

Note: Bold numbers represent the ‘‘Within’’ cluster densities.

calculated between all possible agency connections, densities are also calculated within and between clusters. If interorganizational relationships are fragmented, with a high percentage of relationships between agencies within clusters and few relationships between members of different clusters, then coordination is lacking. If fragmentation is found to be minimal and agencies have relationships with other agencies not in their immediate cluster, then coordination is high (Bolland & Wilson, 1994).6 In the staff interaction example shown in Table 6, it is evident that Cluster III is more connected with other clusters and that Cluster IV is fairly isolated. It would be critical for system planners to examine the agencies included in each cluster to better understand why some clusters are more isolated and whether that is in the best interest of the system.

DISCUSSION AND IMPLICATIONS The primary goal of understanding the providers’ perspectives and conducting the network analysis portion of the participatory evaluation was to examine the coordination of the service system for children and adolescents with SED and their families before and after the implementation of the system-of-care demonstration. Monitoring the status of the system through the providers’ perspective allows for adjustments throughout the implementation to promote a continuous improvement framework. Providers hold a unique position and perspective that is often overlooked, with preference given to system administrators or consumers. Providers must consistently negotiate the system in which they provide services and have intimate knowledge of the strengths and challenges in obtaining the most appropriate array of services for their consumers. In addition, the use of such information values the providers’ perspectives and emphasizes their investment in the system. Finally, gaining perspectives from providers

Evaluating Service System Coordination from the Providers’ Perspective

113

facilitates the evaluation process because providers and system stakeholders are not made to feel as though they are being judged, but rather as part of the process of continually improving the system. One of the most common criticisms of behavioral health service systems is that they are not ‘‘systems’’ but rather fragmented, limited, and poorly coordinated groups of service providers and agencies. One of the stated goals of this initiative was a reduction in fragmentation through the development of a system of care. This chapter illustrated how a network analysis method was used to assess both the structure (NSQ and network analysis) and the effectiveness (ALSS) of the behavioral health services system. Only a subset of the analyses from the study were described here as an illustration of the information that can be gained through this approach; if embraced, this information could serve a powerful role in improving systems. The measures of system effectiveness identiﬁed areas of strength as well as speciﬁc barriers to be overcome. The following measures provided information on the structure of the network relationships: (1) density of the network – how much activity there is among network agencies; (2) centrality – the identity of the most prominent organizations in the network; (3) overall network coordination – the activity that goes on between different clusters of organizations, overcoming the ‘‘cliquishness’’ of relating only to agencies within one’s immediate cluster; and (4) cluster membership – the patterns of agencies that work most strongly together. Because this chapter is intended to illustrate tools that might be applied in participatory health services research, the limitations of the example presented are not discussed extensively. As is true with much applied research, implications of causality associated solely with the implementation of the demonstration system of care must be viewed cautiously since other events during this time period are a threat to validity in this study as in others of system change. Unlike most network studies, baseline data was gathered prior to the implementation of the system of care. But one of the signiﬁcant strengths of this study (baseline data) is challenged by the study’s primary limitation – variation in the size of the networks at baseline and follow-up. The variation in network size resulted from a reduction in the agencies included at followup as well as the participation rate for those agencies. Since the size of the network was reduced by one third at follow-up, the comparability of the some of the density and network structure indicators must be viewed cautiously. Nonetheless, the opportunities presented through the application of network analysis and provider-level assessment in health service research

114

DENINE NORTHRUP

enhance the ability to monitor and detect system changes. The next stage would integrate the system change information with client-level outcomes assessment of those in services at the time of the system change. Multivariate analyses could be conducted to determine the extent to which changes in the system inﬂuence and impact changes in individuals participating in services. Existing research does not support the direct impact of system change on individual outcomes based on randomized and experimental designs (Bickman, 2002; Bickman & Fitzpatrick, 2002). However, if more subtle ﬂuctuations in systems could be tracked effectively, more sensitive assessment might reveal critical factors critical to promoting optimal conditions for enhancing individual outcomes.

NOTES 1. The project was federally funded by the Center for Mental Health Services (CMHS) of the Substance Abuse and Mental Health Services Administration (SAMHSA). 2. The two most common approaches taken by network analysts for identifying the total number and composition of subnetworks are the social cohesion or coherence model and the structural equivalence model. For the purposes of this study, the coherence model was chosen. The principal rationale behind this decision was that the coherence model examines direct relationships between actors. In contrast, the structural equivalence model does not require that agencies within a group/ cluster have direct ties, thus indirect relationships might be presented. Given this study’s concern with issues of service system coordination, it makes more sense theoretically to examine each of these issues on the basis of direct ties with one another. 3. Each section of the ALSS examined a different aspect of the service delivery system, such as system performance, and contained items that speciﬁcally addressed that aspect, such as the use of waiting lists or the availability of high quality treatment. There was also a summary item in each section that provided an overall rating of that aspect of the system. For example, in the performance section, the summary item asked for an overall rating of system performance. 4. A local alcohol and drug center, local psychiatric hospital, and multiservice community agency were included via data imputation. The imputation strategy involved reﬂecting the data using every other agency’s rating about that agency to be a proxy for how the missing agencies may have responded. A similar strategy has been used in similar research. 5. This is usually accomplished through the use of computer-based clustering applications. In this study, the structural organization of the current behavioral health service network was determined by using the coherence procedure with a social network analysis software (NETCLUS: Bolland & Woods, 1987). UCINET V (Borgatti, Everett, & Freeman, 1992) was used for data management and calculation of additional structure and coordination measures.

Evaluating Service System Coordination from the Providers’ Perspective

115

6. The complement of the Somers’ d statistic is used for the coordination index, based on the maximum clustering solution. The coordination index can range from 0 (all activities within clusters and, thus, no coordination) to 1 (complete coordination).

REFERENCES Anthony, W. (2005). Psychiatric rehabilitation and the science of possibilities. Psychiatric Rehabilitation Journal, 28, 313–314. Anthony, W., Cohen, M., & Farkas, M. (2005). The practice of psychiatric rehabilitation: Plans and interventions. In: Larry Davidson, Courtenay Harding & LeRoy Spaniol (Eds), Recovery from severe mental illnesses: Research evidence and implications for practice (Vol. 1, pp. 335–357). Boston, MA: Center for Psychiatric Rehabilitation/Boston U. Bickman, L. (2002). Evaluation of Fort Bragg and Stark County Systems of Care for children and adolescents. American Journal of Evaluation, 23(1), 67. Bickman, L., & Fitzpatrick, J. L. (2002). Evaluation of Fort Bragg and Stark County Systems of Care for children and adolescents: A dialogue with Len Bickman. American Journal of Evaluation, 23(1), 69–80. Bolland, J. M., & Wilson, J. V. (1991a). The Madison County youth services network: The ﬁnal report of the United Way needs assessment project. University of Alabama, Institute for Social Science Research. Bolland, J. M., & Wilson, J. V. (1991b). A network approach to the assessment and development of comprehensive coordination in the delivery of community-based services to the elderly. University of Alabama, Institute for Social Science Research. Bolland, J. M., & Wilson, J. V. (1994). Three faces of integrative coordination: A model of interorganizational relations in a community-based health and human services. Health Services Research, 29(3), 341–366. Bolland, J. M., & Woods, J. A. (1987). Coherence vs. structural equivalence: An evaluation of a new clique-detection algorithm. Paper presented at the 1987 Sunbelt Network Conference, Clearwater Beach, FL. Borgatti, S. P., Everett, M. G., & Freeman, L. C. (1992). UCINET IV. Columbia, SC: Analytic Technologies. Heﬂinger, C. A. (1996). Measuring service system coordination in managed mental health care for children and youth. Evaluation and Program Planning, 19, 155–163. Heﬂinger, C. A., & Northrup, D. A. (1998). Measuring change in mental health services coordination under managed mental health care for children and adolescents. In: J. P. Morrissey (Ed.), Research in community mental Health Volume 10: Social Networks and Mental Health. Stamford, Connecticut: JAI Press. Heﬂinger, C. A., & Northrup, D. A. (2000a). What happens when capitated behavioral health comes to town? The transition from the Ft. Bragg demonstration to a capitated managed behavioral health contract. The Journal of Behavioral Health Services and Research, 27(4), 390–405. Heﬂinger, C. A., & Northrup, D. A. (2000b). Community-level changes in behavioral health care following capitated contracting. Child and Youth Services Review, 22(2), 175–193. Joint Commission on Accreditation of Healthcare Organizations. (2006). Facts about the BHC Standards Supporting Recovery and Resilience. Retrieved May 10, 2006, from http://www.jointcommission.org/AccreditationPrograms/BehavioralHealthCare/Standards/ facts_bhc_recov_resil.htm

116

DENINE NORTHRUP

Knoke, D., & Kuklinski, J. H. (1982). Network analysis. Beverly Hills, CA: Sage Publications. Mizruchi, M. S., & Galaskiewicz, J. (1993). Networks of interorganizational relations. Sociological Methods and Research, 22, 46–70. Morrissey, J. P. (1992). An interorganizational network approach to evaluating children’s mental health service systems. New Directions for Program Evaluation, 54, 85–98. Morrissey, J. P., Calloway, M., Bartko, W. T., Ridgley, S., Goldman, H., & Paulson, R. I. (1994). Local mental health authorities and service system change: Evidence from the Robert Wood Johnson Program on chronic mental illness. The Millbank Quarterly, 72(1), 49–79. Morrissey, J. P., Johnsen, M. C., & Calloway, M. O. (1998). In: M. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: Programs and evaluation of best practices. Austin, TX: PRO-ED, Inc. Morrissey, J. P., Tausig, M., & Lindsey, M. L. (1985). Network analysis methods for mental health service system research: A comparison of two community support systems. (DHHS Publication No. ADM 90-1679). Washington, DC: U.S. Government Printing Ofﬁce. Oliver, A. L., & Montgomery, K. (1996). A network approach to outpatient service delivery systems: Resources ﬂow and systems inﬂuence. Health Services Research, 30, 771–789. Provan, K. G., Isett, K. R., & Milward, H. B. (2004). Cooperation and compromise: A network response to conﬂicting institutional pressure in Community Mental Health. Nonprofit and Voluntary Quarterly, 33, 489–514. Provan, K. G., & Milward, B. H. (1995). A preliminary theory of interorganizational network effectiveness: A comparative study of four Community Mental Health Systems. Administrative Science Quarterly, 40, 1–33. Provan, K. G., & Milward, B. H. (2001). Do networks really work? A framework for evaluating public-sector organizational networks. Public Administrative Review, 61, 414–424. Provan, K. G., Veazie, M. A., Teufel-Shone, N. I., & Huddleston, C. (2004). Network analysis as a tool for assessing and building community capacity for provision of chronic disease services. Health Promotion Practice, 5, 174–181. Stroul, B. (1989). Community support systems for persons with long-term mental illness: A conceptual framework. Psychosocial Rehabilitation Journal, 12(3), 9–26. Stroul, B. A., & Friedman, R. M. (1986, revised). A system of care for severely emotionally disturbed youth. Washington, DC: CASSP Technical Assistance Center at Georgetown University. Van de Ven, A. H. (1976). On the nature, formation, and maintenance of relations among organizations. Academy of Management Review, 1, 24–36. Van de Ven, A. H., & Ferry, D. L. (1980). Measuring and assessing organizations. New York: Wiley.

MEASURING CHILDREN’S SYSTEMS OF CARE USING ANONYMOUS DATA SETS: CASELOAD OVERLAP, SERVICE SYSTEM INTEGRATION, AND NUMBER OF PROGRAMS PER PERSON John A. Pandiani, Christine VanVleck and Steven M. Banks The vision of an integrated, coordinated ‘‘system of care’’ has been central to discussions of meeting the needs of children and adolescents with severe emotional disturbances for years. This concept has helped guide the professional activity of people working with children and adolescents at least since the original publication of ‘‘A System of Care for Children and Youth with Severe Emotional Disturbances’’ (Stroul & Friedman, 1986) two decades ago. Interagency coordination and collaboration is one of the core elements of the system of care philosophy. Based on the recognition that ‘‘no single agency can be effective at serving youngsters with emotional disturbances’’, and that ‘‘child-serving agencies must share responsibility for developing, providing, Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 117–136 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14007-6

117

118

JOHN A. PANDIANI ET AL.

funding, and overseeing service delivery’’, Stroul, Lourie, Goldman, and Katz-Leavy (1992) identiﬁed a variety of mechanisms for ensuring integration and collaboration, but the literature contains scant discussion of methods for measuring the degree to which systems of care are integrated (or segregated). The measurement of service system integration, however, is particularly important because children and adolescents with severe emotional disturbances have challenges that impinge upon multiple life domains and frequently require services from more than one agency (Burns et al., 1995). The pages that follow discuss and demonstrate four basic measures of the functioning of systems of care for children and adolescents. The ﬁrst measure regards the caseload overlap between individual treatment programs or service sectors. The second measure regards the degree of caseload segregation/integration that characterizes a complex system of care. The third measure involves the determination of the number of individuals whose needs are being addressed by only one program or service sector. The ﬁnal measure involves the determination of the number of individuals who are served by speciﬁed numbers of programs within large and complex systems of care. These four measures address issues that are basic to the management of systems of care for children and adolescents. Each of them provides a useful tool for evaluating local systems of care. Together they provide a set of independent but complementary set of tools for measuring and comparing the functioning of systems of care for children and adolescents. This chapter is written within the legal, organizational, and technological context of the ﬁrst decade of the 21st century. The promise of massive integrated databases has yet to be fulﬁlled. The development of large multisector databases has been slowed by privacy concerns and the technology of direct record linkage has proved to be more difﬁcult than anticipated. The statistical technology used in this chapter avoids privacy concerns by using anonymous, HIPAA compliant (Federal Register, 2002), extracts from administrative databases. The utility of the analytical framework developed in this chapter, however, is not limited to analysis of anonymous data sets. This analytical framework also provides a model for measuring basic attributes of systems of care when and if large-scale integrated databases become available.

CASELOAD OVERLAP BETWEEN INDIVIDUAL TREATMENT PROGRAMS OR SERVICE SECTORS A ﬁrst, basic question regarding the complexity of systems of care addresses the degrees to which individuals who are served by one program or service

Measuring Children’s Systems of Care Using Anonymous Data Sets

119

sector are also served by another program or service sector during the same time period. Program managers need to know whether and to what degree the young people on their caseload are also on the caseload of another speciﬁed program. The need for service coordination between community mental health and juvenile justice programs in the same geographical region, for instance, is indicated by the number of young people on both programs. This information also provides higher level system managers with an indication of the degree to which the community mental health program is serving this group of young people who have an elevated need for mental health services. Where there are integrated information systems or where individual information systems share unique person identiﬁers, caseload integration can be directly measured. In the absence of the ability to share unique person identiﬁers, information about caseload overlap may be derived using the statistical technology of Probabilistic Population Estimation (PPE). Probabilistic Population Estimation is a statistical procedure that provides unduplicated counts of the number of children and adolescents who are represented in more than one data set without reference to personally identifying information (Banks & Pandiani, 2001). PPE has three important advantages over alternative approaches. First, the personal privacy of individuals and the conﬁdentiality of medical records are assured because PPE does not depend on information that identiﬁes speciﬁc individuals. Second, because the methodology relies on existing databases, it does not require the commitment of substantial amounts of staff time or ﬁnancial resources required for special purpose data collection. Finally, PPE can support retrospective evaluation of changes in systems of care that have occurred in the past, and provide longitudinal baseline data for evaluating current or anticipated changes in systems of care wherever basic client information resides in electronic databases. Probabilistic Population Estimation allows researchers, policy analysts, and evaluators to answer two basic questions that have frequently remained unanswered because existing data sets lack unique person identiﬁers across organizations and service sectors: ‘‘How many people have contact with a service system?’’ and ‘‘How many people are served by more than one organization, service sector, or service system?’’ PPE provides these estimates by combining information on the distribution of dates of birth in data sets with information on the distribution of dates of birth in the general population to produce valid and reliable estimates of the number of people represented. For example, if 220 dates of birth were represented in a data set that describes all male children served by community mental health programs

120

JOHN A. PANDIANI ET AL.

during 2005 who were born in 1990, PPE would indicate that 338 unique individuals were represented in that data set. Similarly, if a data set with information on male children born in the same year who were on the caseload of the juvenile justice agency during 2005 included 279 dates of birth, that data set would include 529 unique individuals. To determine the number of children and adolescents shared across data sets that do not include a common person identiﬁer, the sizes of three populations are determined, and the results are compared. First, the number of young people represented in each of the original data sets is determined. In this case, the original data sets are the ﬁle that describes all community mental health clients, and the data set that describes all individuals on the juvenile justice caseload during the same year. Second, these two data sets are combined and the number of unique individuals represented in the combined data set is determined. The number of people shared by the two data sets is the number of individuals on the caseload of both the community mental health and the juvenile justice program during 2005. Mathematically, the number of people who are shared by the two data sets is the difference between the sum of the numbers of people represented in the two original data sets and the number of people represented in the combined data set. In terms of mathematical set theory (Whitehead & Russell, 1927), the size of the intersection of two sets (A\B) is the difference between the sum of the sizes of the two sets (A+B) and the size of the union of the two sets (A[B): ðA \ BÞ ¼ A þ B ðA [ BÞ The size of the two original data sets and the size of the combined data set may be determined using the PPE as described above. In the hypothetical example introduced above, there were 220 dates of birth representing 338(7) young people in the mental health data set and 279 dates of birth representing 529(7) individuals in the juvenile justice data set. (The symbol (7) is used in this chapter to represent the statistical uncertainty associated with counts and rates based PPE.) When the two data sets were joined, the combined data set included 316 unique dates of birth. PPE indicates that 736(7) individuals are represented in this combined data set. The overlap between the mental health and the juvenile justice data sets is the difference between the sum of the numbers of people in the two original data sets (8577) and the number of people in the combined data set (7367). In this hypothetical example, 121(7) of the total 338(7) people who had been served by the children’s mental health

Measuring Children’s Systems of Care Using Anonymous Data Sets

121

programs were also served by the juvenile justice program during the same time period. The ability of this statistic to provide valid and reliable estimates (with known conﬁdence intervals) of these basic parameters of service systems is particularly valuable where issues of conﬁdentiality or data quality limit the utility of unique identiﬁers (Pandiani, Banks, & Schacht, 1998). Table 1 provides the rates of caseload overlap between community mental health children’s services programs and juvenile justice programs in 10 regions of one statewide system of care. Because these two programs did not share unique person identiﬁers, PPE was used to measure caseload overlap. As you will see, the proportion of mental health service recipients who were also on the juvenile justice caseload in the same service area ranged from less than 10% in four regions to 20% in one region. The proportion of young people on the juvenile justice caseload that were also on the mental health caseload was even greater, ranging from 24% to 57%. This variation indicates that the degree to which community mental health programs helped meet the needs of this high-risk population varied substantially within the state. Similar analyses would provide measures of caseload overlap between children’s mental health and other relevant human services programs such as special education. Because the data used in this analysis are available for other states and for a number of years, this analysis can be used to examine change over time as well as variation among a broader range of geographical locations. Table 1. Caseload Size and Caseload Overlap by Region Children’s Mental Health and Juvenile Justice Programs Caseloads. Region

Caseload Size

Caseload Overlap

Mental Health Juvenile Justice Number Mental Health (%) Juvenile Justice (%) 1 2 3 4 5 6 7 8 9 10

941713 832712 1,722725 32573 1,306719 994714 51078 707710 55478 712710

14873 32375 765712 12072 38476 24574 15273 28275 28575 22374

8575 11079 256723 2973 122713 11778 3776 7479 11378 10677

971 1371 1571 971 971 1271 771 1071 2072 1571

5774 3472 3371 2473 3272 4872 2473 2672 4072 4873

122

JOHN A. PANDIANI ET AL.

CASELOAD SEGREGATION/INTEGRATION WITHIN A COMPLEX SYSTEM OF CARE As noted earlier, the degree to which child serving agencies share responsibility for children and adolescents has been recognized as an important measure of service system performance for a number of years. A child focused measure of this shared responsibility is provided by the caseload segregation/ integration ratio (CSIR) (Pandiani, Banks, & Schacht, 1999). Caseload segregation/integration has been measured using anonymous records from children’s mental health, child protection, and special education programs on a statewide basis across states (Pandiani, Banks, & Geertsen, 2001). Levels of caseload segregation were found to be related to a number of treatment outcomes on both the individual and the community level (Pandiani, Banks, & Schacht, 2001). Individual level outcomes include incarceration (for boys), maternity (for girls), and hospitalization for behavioral health care (for both genders). Community level outcomes include rates of out-of-home placement, community wide maternity, and hospitalization rates. The CSIR, system integration, is calculated using the following formula: D D CSIR ¼ 1 C 1 100 U LU In this formula, ‘‘D’’ is the sum of the unduplicated counts of children and adolescents served by each sector. Children and adolescents served in more than one service sector are counted more than once in ‘‘D’’. ‘‘U’’ is the unduplicated count of children and adolescents served by any of the three service sectors. ‘‘LU’’ is the unduplicated count of children and adolescents served by the largest service sector. The quantity ‘‘D/U’’ is a raw segregation/integration ratio that is not suitable for comparison across service systems. In order to provide for comparison across local systems of care, this raw ratio is mathematically adjusted by determining the logically possible range of values for any given local service system and expressing the result on a scale that ranges from 0 to 100. Irrespective of the size of the different service sectors, if no individuals are shared by any service sectors, CSIR will produce a value of 0. Similarly, if the caseload of every service sector is completely included by every service sector that is larger (regardless of the number of service sectors in the local system of care), the CSIR produces the value 100. As an example of the calculation of a CSIR, we have used data provided by Burns et al. (1995) that describes the system of care serving children and

Measuring Children’s Systems of Care Using Anonymous Data Sets

123

adolescents with a severe emotional disorder in the Great Smoky Mountain Region of North Carolina. Burns sample included an unduplicated total (U) of 68 children and adolescents who received services from ﬁve different service sectors including education (N ¼ 49), mental health (N ¼ 28), child welfare (N ¼ 11), health (N ¼ 7), and juvenile justice (N ¼ 3). The sum of the numbers served in the ﬁve sectors (D) is 98. The largest service sector (LU) served 49 children and adolescents. The CSIR for this system of care was 44. 98 98 CSIR ¼ 1 C 1 100 68 49 Where unique person identiﬁers are available, CSIR may be calculated using these identiﬁers to derive unduplicated counts of children and adolescents served across service sectors. Where issues of personal privacy or organizational complexity limit the availability of unique person identiﬁers, the unduplicated counts required to calculate the CSIR may be derived using the method of PPE. The following example illustrates the calculation of the CSIR in a setting in which child serving agencies do not share unique person identiﬁers. This example relied exclusively on existing databases maintained by state level agencies representing each of the three distinct service sectors: mental health, juvenile justice, and education. Information on all children and adolescents served by children’s mental health programs in FY2002 was obtained from computer ﬁles maintained by the state mental health authority. Data ﬁles describing all children and adolescents who were on the caseload of the juvenile justice agency during this same period were obtained from that agency. Data ﬁles describing all children and adolescents who were on an Individualized Educational Plan for an emotional behavioral disability was obtained from the state Department of Education. All three data sets included each person’s date of birth and gender, and a geographical code. None of these data sets include a unique person identiﬁer (i.e., social security number, or name and address). Table 2 provides the unduplicated count of young people served in each of the three service sectors, the unduplicated count in the service system as a whole, and the CSIR for each of 10 geographical regions. As you will see, the CSIRs for the majority of the regions fell into the relatively narrow range 28–32, although the complete range of CSIR scores in this state varied from 22 to 57. Clearly the level of caseload segregation/integration varied substantially among regions in this state.

124

Table 2. Region 1 2 3 4 5 6 7 8 9 10

JOHN A. PANDIANI ET AL.

Caseload Segregation/Integration Ratio (CSIR) for 10 Regions of One State FY2002. CSIR 57 30 31 32 38 28 22 24 33 40

THE NUMBER OF INDIVIDUALS SERVED BY ONLY ONE PROGRAM Caseload segregation/integration ratio measures the level of segregation/ integration that characterizes a system of care. CSIR does not, however, provide information regarding the likelihood that children in a system of care will receive services from only one program. Typically, program administrators observe that most of their clients also receive services from many other programs and conclude from this observation that most children in the system of care are served by many, if not most, programs, and that very few are served by only one program. (See the Appendix to this chapter, ‘‘The Administrators’ Misconception’’, for a more detailed discussion of this phenomenon.) While the conclusion that most children in the system of care are served by many programs may in fact be true, it is far from necessarily true. For example, let us consider a local system of care comprised of three programs that each served exactly 300 children during the past year. In this hypothetical system of care, each of the three program directors knows that exactly 60% of their clients were also on the caseload of at least one other program. When the three directors exchange their knowledge regarding shared caseloads, it is not surprising that they collectively conclude that the vast majority of all children in the system of care appear on multiple caseloads.

Measuring Children’s Systems of Care Using Anonymous Data Sets

125

In fact, it is possible that as few as 33% of all children in this system of care were served by more than one program during the year. The vast majority, 67%, could have been served by only one program. How can this be? In this hypothetical example, 180 of the 300 children on each caseload had been served by at least one other program, while 120 of the 300 children on each caseload had not been served by any other programs. If the 180 children who were identiﬁed by the director of the ﬁrst program as being on another caseload were exactly the same 180 children identiﬁed by both of the other program directors as being on another caseload, 67% of all children in the system of care would have been served by only one program during the year. This is true because the unduplicated number of children in this hypothetical system of care is not 900 (300 children in each of three programs). The unduplicated number is actually 540 children. 120 120 120 +180

children children children children

served served served served

only by program 1 only by program 2 only by program 3 by all three programs

540 children in the system of care In fact, the maximum possible number of children that were seen by more than one program in this hypothetical example is 43% (270 of a total of 630 children in the system of care). If this system of care had a functioning integrated information system, and if the program directors had asked the information system for the answer to this question, there would be no misconception. Unfortunately that is rarely the case in human service systems. (The implications of this misconception are also discussed in the Appendix to this chapter regarding ‘‘The Administrators’ Misconception’’.) In the absence of a functioning integrated information system, however, it is still possible to know how many children in a complex system of care are served by only one program, and how many are served by other speciﬁed numbers of programs. In order to determine the precise number of children served by only one program in the absence of a functioning integrated information system, two quantities need to be estimated. One quantity is system-wide. It is the unduplicated count of individuals on the caseload of any program in the system

126

JOHN A. PANDIANI ET AL.

Table 3.

Direct Service Caseload Overlap Children and Adolescents.

Department

Social Welfare Mental Health Corrections Disability Services Health Juvenile Justice Substance Abuse Total Direct Service Caseload

Total Direct Service Caseload

Served by Exactly One Direct Service Department

14,5247110 8,640740 567711 33975 4,112726 5,166724 1,243714

10,7527105 4,766782 268724 152715 2,549755 2,514758 527732

74%70.8% 55%71.0% 47%74.3% 45%74.4% 62%71.4% 49%71.1% 42%72.6%

27, 4967110

21,5297161

78%70.7%

of care. The second quantity is program speciﬁc. It is the unduplicated number of individuals on the caseload of all programs but the speciﬁed programs. Both of these unduplicated counts can be determined using PPE (described above). PPE is a statistical procedure for estimating the unduplicated number of people represented in a data set based on the distribution of dates of birth and genders in the data set. The number on the caseload of only program 1, for instance, is the difference between the total numbers served in the system of care and the number on the caseload of any program other than program 1. The number on the caseload of only one program is the sum of the numbers of individuals who are only on the caseload of each individual program in the system of care. To demonstrate this technique in a real world system of care, seven major programs that provide direct services to children and adolescents in one state were selected for analysis. These programs include mental health, health, substance abuse, disability services, juvenile justice, social welfare, and correctional programs. Statistical analysis indicated that 78% of the young people in this system of care had received direct services from only one program. This is true despite the fact that more than half of the young people served by four of the seven programs had been served by more than one program (see Table 3).

THE NUMBER OF INDIVIDUALS SERVED BY SPECIFIED NUMBERS OF PROGRAMS The procedure described above provides valid and reliable estimates of the number of young people served by only one program, but it does not provide

Measuring Children’s Systems of Care Using Anonymous Data Sets

127

information about the number of young people served by exactly two programs, or all seven programs, or any other speciﬁc numbers of programs. In order to determine the number of young people served by 1, 2, 3, etc. programs, we have developed four discrete procedures. The ﬁrst procedure (described above) involves estimating the number of individuals served by only one program. The second procedure involves estimating the number served by exactly two programs. The third procedure involves estimating the maximum possible number of unique young people who were served by all programs, and in this example by exactly six, and exactly ﬁve programs. The fourth and ﬁnal procedure involves estimating the number served by each of the remaining numbers of programs (exactly three programs and exactly four programs, in this example). Each of these procedure involve different statistical analyses. Procedure One is described in the preceding section. Procedure Two. The determination of the number of individuals on the caseload of exactly two programs requires an elaboration of Procedure One. This elaboration involves the construction of three data sets that relate to each possible pair of programs in the system of care. The ﬁrst data set is the caseload of one of the programs (Children’s Mental Health, for instance). The second data set is the caseload of a second program (Juvenile Justice, for instance). The third data is the combined caseload for all other programs. PPE is used to determine the number of individuals who are on the caseload of both of the speciﬁed programs (social welfare and mental health) but are not on the caseload of any other program. A graphic depiction of this situation is provided in Fig. 1. Two of the circles represent the caseloads of individual programs (social welfare and mental health in this example). The third circle represents all other programs. In this ﬁgure, the area indicated by the asterisk () is the number of young people who are served by both mental health and social welfare but are not served by any other program. This result is based on the logic of classical set theory as described above with regard to the overlap between two data sets. The identiﬁed area can be expressed mathematically as: ðA \ BÞ ðA \ B \ CÞ The quantity ðA \ B \ CÞ can be expressed as: A þ B þ C ðA [ BÞ ðA [ CÞ ðB [ CÞ þ ðA [ B [ CÞ. Therefore, the identiﬁed area may be expressed as: C þ ðA [ CÞ þ ðB [ CÞ ðA [ B [ CÞ

128

JOHN A. PANDIANI ET AL.

All Other Programs

Program One

* Program Two Fig. 1.

Hypothetical 3-Way Overlap.

The important observation is that we can estimate the size of identiﬁed area by estimating the size of three combinations of data sets. The ﬁrst combination is mental health and all other programs (A[C). The second combination is juvenile justice and all other programs (B[C). The third is the combination of mental health, juvenile justice, and all other programs (A[B[C). The number of young people on the caseload of exactly two programs is the sum of the numbers of individuals who were on the caseload of both of each possible pair of programs, but were not on the caseload of any other program. In this real world example, with seven different programs, it was necessary to estimate the unduplicated number of young people on each of 21 possible pairs of programs. In this system of care, 15% of all children received services from exactly two different programs. Procedure Three. Although, conceptually, Procedure Two could be used to estimate the number of individuals on the caseload of exactly three programs, exactly four programs, etc., the statistical uncertainty associated with these estimates becomes unacceptably large. For this reason, a different procedure was developed to estimate the maximum possible numbers of individuals on the caseload of all seven programs. This procedure is based on the observation that if an individual were on the caseload of all seven programs, his or her date of birth and gender would appear in the caseloads of all seven programs. We estimate the maximum possible number of unique individuals on the caseload of all seven programs by applying PPE to the number of unique date of birth and gender combinations that are

Measuring Children’s Systems of Care Using Anonymous Data Sets

129

represented in every program. This provides the maximum possible number of individuals on the caseload of all seven programs. It is an overestimate of the true value to the degree that different individuals with the same date of birth and gender are served by different programs in the system of care. In the current example, there was no date of birth and gender combination observed in each of the seven databases. Therefore, no individual was seen by all seven programs within the year. The procedure to estimate the maximum number of unique individuals in all seven programs may be used to estimate the maximum number of unique individuals in fewer programs. For example, if an individual is in six of the seven programs, his or her date of birth would appear in six of the seven databases. Thus, we estimate the maximum possible number of unique individuals on the caseload of six of the seven programs by applying PPE to the number of unique date of birth and gender combinations that are represented in six of the seven databases. As before, this will overestimate the true value to the degree that different individuals with the same date of birth and gender were served by different programs in the system of care. Conceptually, this procedure can also be repeated for every combination of programs, however, as in the second step, at some point, the statistically uncertainty would again become unacceptably large. For this reason, a fourth step in the procedure is used to estimate the number of individuals in exactly ‘‘n’’ programs that could not be estimated using steps 1 and 2 (which estimates exactly one and exactly two programs) and step 3 (which estimates exactly seven, and in this example exactly six, and exactly ﬁve). Procedure Four. In order to determine the number of individuals served by the remaining numbers of programs, we apply a mathematical procedure known as linear programming. Linear programming is a method for determining the maximum (or minimum) value for a speciﬁc quantity given a series of equations and constraints regarding relationships among the variables in the system (Dantzig & Thapa, 1997). In this case the constraints are provided by the results provided by Procedures 1, 2, and 3, above. Procedures 1 through 3 provided estimates of the number of individuals served by only 1, exactly two, and exactly ﬁve, six, and seven programs (and the conﬁdence intervals for each). In order to determine the number of individuals served by exactly three and exactly four programs, we begin with two mathematical observations. First, if we denote ‘‘a’’ as the number of individuals in only 1 program, ‘‘b’’ as the number of individuals in exactly two programs, through ‘‘g’’ the number of individuals in exactly seven programs, the unduplicated count of individuals is the sum of ‘‘a’’ through ‘‘g’’. Second, the duplicated count of individuals is the sum of

130

JOHN A. PANDIANI ET AL. Overall Direct Service Caseload Complexity

100%

Percent of Clients

75%

50%

25%

0% One

Two

Three

Four

Five

Six

Seven

Number of Departments

Fig. 2.

Overall Direct Service Caseload Complexity.

1a+2b+3c+4d+5e+6f+7g. Substituting the known values for a, b, e, f, and g, linear programming provides the minimum and maximum possible number of individual who were served by exactly three and exactly four programs. In this example, the number of individuals served by exactly three programs is about 200 (0.8% of the 27,496 unique children and adolescents) and the number of individual served by exactly four programs is about 13 (0.05% of the 27,496 unique children and adolescents) (Fig. 2).

DISCUSSION This chapter has provided a basic set of quantitative measures of service system performance designed speciﬁcally to provide information about the functioning of complex systems of care for children and adolescents. The examples provided in this chapter have focused on one statewide system of care and on variation among local systems of care within that state. This

Measuring Children’s Systems of Care Using Anonymous Data Sets

131

approach can also be used to examine and monitor service system performance over time and to compare patterns of care provided to young people with differences in clinical and demographic characteristics. Longitudinal analyses can contribute to our understanding of causes and consequences of changes in systems of care. Cross-sectional analysis can help us to identify disparities in the way in which young people are served. Together with measures of system performance that focus on treatment outcomes, the measures of system level practice patterns described in this chapter have the potential to improve the ways in which systems of care address the needs of young people with emotional and behavioral disorders. The approach to service systems research presented in this chapter is based on the belief that the large administrative and operational databases that our society generates on an ongoing basis can contribute to an unprecedented growth of knowledge in medical and behavioral sciences (Pandiani & Banks, 2003). This chapter was designed to contribute to the development of new methodologies that will allow us to learn from our massive stores of data. These data include public health, vital records, social and criminal justice programs, public and private insurance claims databases. They also include databases that indicate positive participation in society, such as school participation and performance, and gainful employment. We live in an information-rich society. We have a responsibility to use this resource to promote the advancement of knowledge and the quality of care that we provide to young people. Our administrative and operational databases have the potential to move the management of systems of care forward at a rapid rate. In the past, program administration tended to operate at the clinical level, seeking to understand the problems and needs of individual young people. The approach demonstrated in this chapter is designed to provide program administrators with a data based system level perspective on the functioning of our systems of care. Our wealth of administrative and operational data has the potential to provide program managers with the information they need to design, implement, and evaluate systems of care that more effectively and more efﬁciently meet the needs of the young people they serve and society as a whole. In order to actualize this potential, human service systems need to adopt new models for creating and sharing knowledge that embrace the information stores and analytical power that we now have at our ﬁngertips.

132

JOHN A. PANDIANI ET AL.

APPENDIX: THE ADMINISTRATORS’ MISCONCEPTION Program administrators tend to rely on direct experience with staff and clients to arrive at their understanding of service system characteristics. This chapter has demonstrated how analysis of existing databases using appropriate statistical techniques can provide objective measures of service system structure and complexity. This Appendix provides a comparison of one subjective ‘‘estimate’’ of caseload overlap (a basic characteristic of a complex system of care) and the estimate provided by statistical analysis of administrative databases. In addition, this Appendix includes a discussion of ways to understand the difference between estimates based on direct experience and those based on analysis of large administrative data sets. Two Estimates of Caseload Complexity

100% Researchers' Estimates Administrators' Estimates

Percent of Clients

75%

50%

25%

0% LT 4

4+

Number of Departments

Measuring Children’s Systems of Care Using Anonymous Data Sets

133

In this example, a group of program managers, based on their collective experience, estimated that, ‘‘Some 70 percent of clients receive, or are eligible for, services from four or ﬁve separate departments or divisions of the total 7 major programs in the system of care’’. ‘‘Clients with multiple needs may have to visit several ofﬁces, tell their stories repeatedly and contend with repetitious (and sometimes contradictory) requirement and case plans’’. A variety of approaches to addressing this problem were proposed. Where managers had believed that 70% of their clients were served by four or ﬁve departments, statistical analysis using the techniques described in this chapter indicated that fewer than 2% received direct services from four or more departments. In fact, 78% of their clients had received direct services from only one department. Clearly, the program administrators’ estimate described a much more complex caseload than actually existed. More important, the policy implications and the resources required to address a problem that affects 1 or 2% of a caseload are very different from those necessary to address a problem that affects 70% of the caseload. We describe the difference between the perspective of program administrators and the results of statistical analysis as the results of an ‘‘administrator’s misconception’’. We believe this administrator’s misconception can be understood from at least two perspectives: a mathematical perspective, and a perspective that focuses on human service information systems. In either case, the implications for the design and management of a system of care are similar.

MATHEMATICAL CONSIDERATIONS Some of the difference between the administrators’ and the statisticians’ estimates is related to the fact that the administrators’ sample (the cases the administrator thinks of when arriving at a conclusion about the caseload) may not be representative of the caseload as a whole. As Cohen and Cohen (1984) have observed, there is a strong tendency for clinicians (and, we would add, administrators) to remember ‘‘trouble’’ cases, cases that are difﬁcult from a treatment perspective and that remain on the caseload for a long period of time. These cases are much more likely to be involved with multiple programs. For this reason, the administrator’s mental ‘‘sample’’ is likely to include a much greater proportion of individuals with involvement with multiple programs than the total caseload studied by the statisticians. The statistical analysis was applied to all cases, including the simple, shortterm cases that are more likely to be served by only one program.

134

JOHN A. PANDIANI ET AL.

A second mathematical basis for the administrator’s misconception is the ironic tendency for the percent of clients who are on the caseload of multiple programs to decrease as the number of programs under consideration increases. This occurs because clients who are on the caseload of multiple programs may be counted multiple times by administrators who pool their perceptions of their individual caseloads. For example, if two programs serve 100 clients each, and 50% of each caseload was shared with the other, the total number served would be 150 (50 only served by the ﬁrst program, plus 50 only served by the second program, plus 50 served by both programs). The 50 shared would be only 33% of the total 150 served.

INFORMATION SYSTEMS Contemporary public sector information systems are not designed to answer the kind of macro-organizational level questions that are being addressed here. Current information systems tended to be developed at the program level during a period of information system proliferation when having your own information system was a status symbol that provided the power to manage information ﬂow. In many programs, external (e.g., federal) demands for information and funding of information system development ﬂowed to program level units. The result of these factors is fragmented and incompatible information systems. Cross data set analysis is also hampered by increasing limitation on the sharing of personally identifying information. Conﬁdentiality regulations (e.g., IRB regulations and HIPAA) severely limit the ability of traditional information systems and analytical approaches to answer the questions raised here.

IMPLICATIONS These ﬁndings demonstrate that program administrators’ subjective perceptions of organizational complexity (as measured by client participation in multiple treatment programs) can include substantial overestimation of the number of service recipients with involvement in multiple treatment programs. We believe this overestimation can have serious consequences for management decisions, including clinical stafﬁng levels necessary to coordinate treatment across programs, and information technology necessary to manage complex cases. Exclusive reliance on subjective information or

Measuring Children’s Systems of Care Using Anonymous Data Sets

135

incomplete information can have serious consequences for management decisions, especially regarding the organization, reorganization, and management of human service programs. This is not to say that administrators’ subjective perceptions have no role in the assessment of organizational structure and functioning. Program administrators’ subjective perceptions should play a key role in the identiﬁcation of values that guide the system of care and the application of those values to the interpretation of quantitative indicators. In this case, such discussions could focus on the relative value of specialized vs. generalized services for young people, and the individual child vs. the family unit as the focus of intervention, for instance. This discussion of human service values, however, should be informed by quantitative indicators that can be provided by innovative statistical techniques for extracting important quantitative indicators from existing disconnected databases. This information regarding the structure and functioning of complex systems of care is one of the key elements in understanding and improving our systems of care for children, adolescents, and their families.

REFERENCES Banks, S. M., & Pandiani, J. A. (2001). Probabilistic population estimation of the size and overlap of data sets based on date of birth. Statistics in Medicine, 20, 1421–1430. Burns, B. J., Costello, E. J., Angold, A., Tweed, D., Stangl, D., Farmer, E. M. Z., & Erkanli, A. (1995). Children’s mental health service use across service sectors. Health Affairs, 14, 147–159. Cohen, P., & Cohen, J. (1984). The clinicians illusion. Archives of General Psychiatry, 41, 1178–1182. Dantzig, G. B., & Thapa, M. N. (1997). Linear programming 1: Introduction. New York: Springer Verlag. Federal Register. (August 14, 2002). Rules and regulations (Vol. 67, Number 157, pp. 53181–53273, speciﬁcally p. 53235). From the Federal Register online via GPO access [wais.access. gpo.gov]. Pandiani, J. A., & Banks, S. M. (2003). Large data sets are powerful. Psychiatric Services, 54(5), 745. Pandiani, J. A., Banks, S. M., & Geertsen, D. C. (2001). Caseload segregation/integration in Utah and Vermont. In: C. Newman, C. Liberton, K. Kutash & R. M. Freedman (Eds), The 13th annual conference proceedings, a system of care for children’s mental health: Expanding the research base (pp. 343–345). Tampa, FL: University of South Florida, The Louis de la Parte Florida Mental Health Institute, Research and Training Center for Children’s Mental Health (http://www.fmhi.usf.edu/institute/pubs/pdf/cfs/rtc/ 13thproceedings/Chapter10.pdf).

136

JOHN A. PANDIANI ET AL.

Pandiani, J. A., Banks, S. M., & Schacht, L. M. (1998). Personal privacy vs. public accountability: A technological solution to an ethical dilemma. Journal of Behavioral Health Services and Research, 25(3), 300–311. Pandiani, J. A., Banks, S. M., & Schacht, L. M. (2001). Caseload segregation/integration and service delivery outcomes for children and adolescents. Journal of Emotional and Behavioral Disorders, 9(4), 232–238. Pandiani, J. A., Banks, S. M., & Schacht, L. S. (1999). Caseload segregation/integration: A measure of shared responsibility for children and adolescents. Journal of Emotional and Behavioral Disorders, 7(2), 66–71. Stroul, B. A., & Friedman, R. M. (1986). A system of care for children and youth with severe emotional disturbances (Rev. ed.). Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Stroul, B. A., Lourie, I. S., Goldman, S. K., & Katz-Leavy, J. W. (1992). Profiles of local systems of care for children and adolescents with severe emotional disturbances (Rev. ed.). Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Whitehead, A. N., & Russell, B. (1927). Principia mathematica (2nd ed., Vol. 1, pp. 211–212). Cambridge: Cambridge University Press.

THE EVIDENCE FOR HOME AND COMMUNITY-BASED MENTAL HEALTH SERVICES: HALF FULL OR HALF EMPTY OR CREATE OTHER GLASSES? Tracy J. Pinkard and Leonard Bickman INTRODUCTION Two major reform movements have shaped child and adolescent mental health services over the past quarter-century: the Systems of Care movement, and more recently, the movement toward evidence-based practice. Results from several studies indicate that youth served in traditional residential or inpatient care may experience difﬁculty re-entering their natural environments, or were released into physically and emotionally unsafe homes (Bruns & Burchard, 2000; President’s Commission on Mental Health, 1978; Stortz, 2000; Stroul & Friedman, 1986; U.S. Department of Health and Human Services, 1999). The cost of hospitalizing youth also became a policy concern (Henggeler et al., 1999b; Kielser, 1993; U.S. Department of Health and Human Services, 1999). For example, it is estimated that from the late 1980s through 1990 inpatient treatment consumed nearly half of all expenditures for child and adolescent mental health care although the services were found not to be very effective (Burns, 1991; Burns & Friedman, 1990). More recent Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 139–178 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14008-8

139

140

TRACY J. PINKARD AND LEONARD BICKMAN

analyses indicate that at least 1/3 of all mental health expenditures for youth are associated with inpatient hospitalization (Ringel & Sturm, 2001). Several early policy papers also comment on the inadequacy of traditional outpatient services (Friedman, 1984; Knitzer, 1982; U.S. Congress Ofﬁce of Technology Assessment, 1986). These services (generally ofﬁce visits) were often available only through private providers and costs were not covered by insurance. Additionally, many services were geographically located in areas difﬁcult for families to access given time and ﬁnancial constraints. Finally, the offered services were fragmented and uncoordinated resulting in a number of youth slipping through the cracks of the service system. In 1984, the Child and Adolescent Service System Program (CASSP) was created by the National Institute of Mental Health (NIMH) in an effort to address these problems. Currently housed under the Center for Mental Health Services (CMHS) of the Substance Abuse and Mental Health Services Administration (SAMHSA), the CASSP initiative was designed to aid states and communities in establishing Systems of Care – ‘‘a comprehensive spectrum of mental health and other necessary services which (sic) are organized into a coordinated network to meet the multiple and changing needs of children and adolescents with severe emotional disturbances and their families’’ (Stroul & Friedman, 1994, p. xx). Broader than continuums of care, which typically only describe mental health services, Systems of Care encompass services as well as ‘‘mechanisms, arrangements, structures, or processes to ensure that the services are provided in a coordinated, cohesive manner’’ (Stroul & Friedman, 1994, p. 20). The Systems of Care philosophy adopts a public health approach that places the responsibility for care with the community as a whole rather than a single agency (Foster & Connor, 2005). This approach centers on several core values and principles that maintain that treatment should be coordinated, community-based, involve families in planning and delivery, and be culturally sensitive (Stroul & Friedman, 1994). The second movement to inﬂuence the current state of Child and Adolescent Mental Health Services (CAMHS) is the development of evidencebased practice. Results of large scale Systems of Care studies showed that coordinating services improved treatment access but not outcomes (Bickman, 1996a, 1996b; Bickman et al., 1994, 1995). As such, there has been a shift in emphasis from the system in which services are delivered to studying the effectiveness of the services themselves (Hoagwood, Burns, Kiser, Ringeisen, & Schoenwald, 2001). The ﬁeld of CAMHS has begun to create a foundation for evidence-based practice. Here evidence-based practice is used to describe several terms including evidence-based service, empirically supported therapies, empirically validated treatments, and so forth.

The Evidence for Home and Community-Based Mental Health Services

141

Stimulated by Cochrane’s 1972 book Effectiveness and Efficiency: Random Reflections on Health Services in the ﬁeld of epidemiology, the evidencebased practice movement ﬁrst took hold in the medical ﬁeld (evidence-based medicine). As deﬁned by the Institute of Medicine (Institute of Medicine, 2001) evidence-based practice is based on the best research evidence available in addition to clinical experience and is consistent with patient values. However as deﬁned by Hoagwood et al., (2001): In the ﬁeld of [child and adolescent] mental health services research, the term ‘‘evidencebased practice’’ refers to a body of scientiﬁc knowledge about service practices... or about the impact of clinical treatments or services on the mental health problems of children and adolescents. (p. 1179)

Although in the past, empirically supported treatments were considered those shown to be efﬁcacious (Chambless et al., 1996), current evidencebased practice means not only conducting studies in controlled laboratory settings to demonstrate that speciﬁc services could work, but also conducting well-designed studies in real-world settings with clients similar to those the typical practitioner would treat (effectiveness studies; Chambless & Hollon, 1998; Hoagwood et al., 2001; Shirk, 2004). Moreover, there is an increasing emphasis on transportability research that addresses the implementation of services, and dissemination research. Together, the Systems of Care movement and the move toward evidencebased practice placed pressure on the ﬁeld of child and adolescent mental heath services not only to provide care within the natural environment of the youth, but also to ensure that the services provided are coordinated and are effective in terms of clinical outcomes, cost and other factors (Huang et al., 2005). This chapter will provide an overview of the most common home and community-based services for children and adolescents provided in Systems of Care, examine the evidence-base for the effectiveness of these services, discuss the challenge of providing evidence-based services when the evidence-base is not very strong, and propose an additional approach to the effectiveness debate.

WHAT ARE HOME AND COMMUNITY-BASED MENTAL HEALTH SERVICES? Home and community-based mental health services (HCBMHS) are those mental health services that are provided for youth with serious emotional disturbances (SEDs) within the youth’s natural environment (home, school,

142

TRACY J. PINKARD AND LEONARD BICKMAN

neighborhood, etc.) These services are guided by: ‘‘an explicit recognition of the importance of ecology and context, recognition of the multiple needs of youth and their families and multifaceted approaches to addressing those needs, a concerted focus on strengths and resources, individualized treatment planning and intervention based on these strengths and needs, and active involvement of the community and focus on building or maintaining community connections for the child and family.’’ (Farmer, Dorsey, & Mustillo, 2004, p. 857)

As Burns (2003) notes, most community-based services are provided by parents, volunteers, and by counselors who do not have formal clinical training but are trained and supervised by mental health professionals. In 1981 legislation was passed that enabled states to offer through the Medicaid Home and Community-Based Services waiver (1915c) program, health and mental health services not previously available through Medicaid programs as well as services to people in their own homes and communities rather than in hospitals or other residential settings (Centers for Medicare and Medicaid Services, 2004). Immediately following in 1982, the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA)/‘‘Katie Beckett’’ waiver was passed. Katie Beckett’s parents successfully petitioned the Federal government to allow their daughter to receive Medicaid funded services in the home setting rather than in a hospital. Under this provision children not traditionally eligible for Medicaid (generally due to family income), but at risk of placement or in placement in a residential treatment facility or who needed at least one emergency stay in a psychiatric hospital in the past 2 years are allowed to use whatever Medicaid services are available in that state, in contrast to only traditional residential treatment. The services included as HCBMHS vary by source. Drawing from Medicaid guidelines, therapeutic foster care (TFC), crisis services, and respite care are examples of recognized HCBMHS (Centers for Medicare and Medicaid Services, 2006). According to Burns (2003), intensive case management (ICM), multi-systemic therapy (MST), treatment foster care, mentoring, family education and support, and wraparound are all considered home and community-based mental health services for children and adolescents. In a careful and comprehensive review of home and community-based services, Farmer et al. (2004) includes community-based residential treatments (including treatment foster care and group homes), multi-modal treatments (i.e., MST), coordination and facilitation services (intensive case management and wraparound), and auxiliary services (family education and support, mentoring, and respite services). Huang et al. (2005) listed HCBMHS as TFC, wraparound care, intensive home-based care, day treatment, mentoring, and respite care.Table 1 provides an overview of the different types of home and

Service

Community-based residential

Therapeutic foster care

Therapeutic group homes

Multi-systemic therapy

Crisis services

Description

Youth, foster family, and natural family

Youth

Youth, family, and community

Youth and family

143

‘‘A service which [sic] provides treatment for troubled children within the private homes of trained families. The approach combines the normalizing inﬂuence of family-based care with specialized treatment interventions, thereby creating a therapeutic environment in the context of a nurturant family home’’ (Stroul & Friedman, 1986, p. 13). ‘‘Group homes are typically small, community based facilities that rely on community resources such as public schools and recreational facilities to provide certain services’’ (Curtis, Alexander, & Lunghofer, 2001, p. 379). ‘‘A comprehensive and individualized treatment approach that addresses the multiple determinants of identiﬁed youth with family problemsyMST targets contributing factors at individual, family, peer, school, and community levels’’ (Henggeler et al., 1999a, p. 174). ‘‘The goals of crisis services include intervening immediately, providing brief and intensive treatment, involving families in treatment, linking clients and families with other community support services, and averting visits to the emergency department or hospitalization by stabilizing the crisis situation in the most normal setting for the adolescent’’ (U.S. Department of Health and Human Services, 1999, p. 178).

Primary Intervention Target(s)

The Evidence for Home and Community-Based Mental Health Services

Table 1. Overview of Home and Community-Based Mental Health Services.

144

Table 1. (Continued ) Service

Auxiliary and supportive services

Family education and support

Mentoring

Coordination and facilitation

Intensive case management Wraparound

‘‘A broad range of interventions intended to educate parents about their child’s disability or train them to manage typical problems, often in combination with social support or group education’’ (Friesen et al., 2005, p. 111). ‘‘A non-professional with good child relationship skills helps children increase their engagement in school or in the communityy’’ (Burns, 2003, p. 968). ‘‘The provision of temporary care to persons with disabilities, with the primary purpose of providing relief to caregivers’’ (Warren & Cohen, 1985, p. 66). ‘‘A commonly used strategy for increasing access to and coordination of servicesy’’ (Farmer et al., 2004, p. 866). A service philosophy: ‘‘a deﬁnable planning process that involves the child and family and is designed to provide a unique set of individualized community services and natural supports based on a set of core principles and values’’ (Farmer et al., 2004, p. 867).

Primary Intervention Target(s) Youth and family

Youth

Family and youth

Youth and family

Youth, family, and community

TRACY J. PINKARD AND LEONARD BICKMAN

Respite care

Description

The Evidence for Home and Community-Based Mental Health Services

145

community-based services for children and adolescents discussed in this chapter. We want to note here the impreciseness of our language. Although the ﬁeld does not draw a sharp distinction between services and treatment, the two are not synonymous. In general we will consider services and settings as synonymous such that a service is generally deﬁned as the setting in which a speciﬁc treatment is provided. Thus calling some service a HCBMHS simply informs us that the intervention occurs in the home or community. In contrast, the term ‘treatment’ should be reserved for speciﬁc interventions, be they medication algorithms or manualized psychotherapy treatments. ‘‘Treatments can be more or less intensive, comprehensive, intrusive, and effective. Treatment can be delivered in a variety of settings. It is the setting that varies with respect to restrictiveness, not the treatmentyit is quite possible that the nature and effect of a particular treatment changes when deployed in different settings; this is an empirical questiony.’’ (Schoenwald, 2002, p. 96)

Treatments include speciﬁc behaviors (not values) that the clinician should use in treatment. Treatments require training and typically a manual that describes these behaviors. The treatment should describe key theoretical constructs, clinical change mechanisms, and therapist interventions. They should be well-articulated models of clinical intervention that are based on scientiﬁc evidence. Within HCBMHS are such treatments as MST (Henggeler, Schoenwald, Rowland, & Cunningham, 2002), and Family Functional Therapy (FFT; Sexton & Alexander, 2003).

EVIDENCE OF THE EFFECTIVENESS OF HOME AND COMMUNITY-BASED MENTAL HEALTH SERVICES ‘‘I am not concluding that services are ineffective – only that there is no systematic evidence that services are effective.’’ (Bickman, 1999, p. 968)

Despite the call for providing evidence-based services, the research base supporting home and community-based services is not strong. Indeed, even the Surgeon General’s Report states that ‘‘evidence for the beneﬁts of some of these services is uneven at best’’ (U.S. Public Health Service, 2000, p. 172); not much has changed since the publication of that report. Much of the evidence is based on studies using a simple pre-post design. Although these designs are useful for ‘‘assessing the likely usefulness of a more rigorous evaluation, searching for promising variables related to success in the program, and preparing an agency for more controlled evaluations in the futurey Simple-group designs cannot answer many questions that are

146

TRACY J. PINKARD AND LEONARD BICKMAN

critical to stakeholders because many interpretations of ﬁndings remain plausible. (Posavac & Carey, 2003, p. 172)

Moreover, pre-post designs with only two data points make it difﬁcult to interpret change (Lambert, Doucette, & Bickman, 2001a). Additionally, with few notable exceptions (e. g., MST) what transpires between the counselor/therapist and client and family in the home or the community is not known. As noted above, other than those services that include some form of manualized clinical treatment, there is little speciﬁcity beyond the location of the services and what is charged for those services. Even with the presence of manual, it is difﬁcult to document what is in the black box of treatment. Privacy and feasibility generally do not permit the videotaping of sessions in the real world. Even if such videotapes were available unless the therapist is following a protocol, it would be difﬁcult to interpret the video. If recording a session is not feasible then we must depend on client and clinician recall of what transpired. Moreover, while the presence or absence of a treatment feature might be able to documented, the judgment of quality of implementation is much more difﬁcult. If it is nonmanualized treatment then just developing a system to describe what occurred in treatment is a major challenge. With a manualized treatment there is at least the possibility of describing the treatment. Included in this chapter is a summary of the results of the excellent Farmer et al. (2004) review for each type of HCBMHS. Additionally, if any signiﬁcant studies were excluded from the Farmer et al. review or if more recent studies have been conducted they will be brieﬂy discussed as well. Farmer et al. conducted an extensive assessment of peer-reviewed published studies of several home and community-based mental heath services for children and adolescents. The authors evaluated each study in terms of research design, the use of a comparison group, replications from more than one team of investigators, and ‘‘readiness for transportability or dissemination and applicability to real world settings’’ (Farmer et al., 2004, p. 859). They also followed Kazdin’s (2004) suggestion of using a continuum of evidence that includes: ‘‘(1) not evaluated, (2) evaluated but unclear (no or possibly negative effects that this time), (3) promising (some evidence), (4) well established (parallel to well-established in conventional schemes), and (5) better/best treatments (treatment shown to be more effective than other evidence-based treatments)’’ (Farmer et al., 2004, p. 859). The rating system used here is also based on Kazdin’s system (2004, 2006) as well as numerous other sources such as the Coalition for Evidence-Based Policy (2003), Society for Prevention Research (2004), etc. While the details of the rating

The Evidence for Home and Community-Based Mental Health Services

147

systems used by different agencies and organizations vary widely, the overarching criteria are generally the same and focus on study design, number of studies, sample speciﬁcation, ﬁndings, replications, and availability of manuals (Kazdin, 2004, 2006; Society for Prevention Research, 2004). Details of the rating system used here are provided in Table 2. Community-Based Residential Treatment Services Community-based residential treatment services are mental health services that require a youth to be removed from the home for some period of time. However, unlike traditional residential services, these services are offered within the local community of the youth. Therapeutic Foster Care Therapeutic foster care (also called specialized foster care, multi-dimensional treatment foster care, treatment foster family care, parent–therapist programs, and family-based treatment) is considered to be the least restrictive form of out-of-home treatment (U.S. Department of Health and Human Services, 1999) and in the case of youth with aggression problems, is often considered the last alternative before placement in a secure facility (Hahn et al., 2005). In 1999, there were 23,000 youth in TFC (Curtis et al., 2001). Like most services, the speciﬁc structure and delivery of TFC varies by program (Farmer, Burns, Dubs, & Thompson, 2002). Most models of TFC are driven by social learning theory that suggests that ‘‘interactions among family members shape prosocial as well as antisocial behavior patterns y therapeutic foster care provides structured and supportive parenting for youth whose parents are unable to do so’’ (Hahn et al., 2005, pp. 73–74). TFC generally is provided to youth with SED by foster parents trained to provide therapeutic intervention (Fisher, Ellis, & Chamberlain, 1999). The training of the foster parents distinguishes this service from traditional foster care in which children and adolescents with SED have not been found to fair so well and, thus, combines aspects of residential treatment with those of a normalized environment. Speciﬁcally, foster parents receive pre-service and in-service training in behavior management and treatment as well as ongoing support, crisis intervention services, and performance evaluations (Hahn et al., 2005). The foster parents are central to the child’s treatment team and play a large role in developing and implementing the youth’s treatment plan. Therapeutic foster parents generally care for only one or two youth at a time and receive more monetary compensation than

148

Table 2.

Study design

Sample speciﬁcation

Signiﬁcant outcome differences Independent replication Manual available Other

Not Evaluated

Evaluated but not Clear

Promising

Well-Established

Better/Best Service

Primarily descriptive studies None in past decade Related to but not speciﬁcally SED Not evaluated

Pre/post or no control group

Quasi-experimental or RCT

Mostly RCT; some quasi-experimental

Mostly RCT; some quasiexperimental

Less than two in past decade Related to but not speciﬁcally SED

More than two in past decade Speciﬁc to SED

Five or more in past decade Speciﬁc to SED

Five or more in past decade Speciﬁc to SED

Negative or equivocal

Majority positive

Majority positive

Majority positive

None

None

None

Yes

Yes

No

No

No

Yes

Yes Shown more effective than other wellestablished service

TRACY J. PINKARD AND LEONARD BICKMAN

Number of studies

Description of the Evidence Rating System.

The Evidence for Home and Community-Based Mental Health Services

149

do traditional foster parents (Fisher et al., 1999). Children typically remain in the foster home for 6–9 months or longer (Burns, 2003; Hahn et al., 2005). Besides treatment for the youth, and training and support for the foster family, TFC service models also include the natural parents in treatment (Evans, Armstrong, Kuppinger, Huz, & Johnson, 1998a). Providing family therapy may improve family functioning so that if and when the youth returns to the natural family, they return to a supportive environment (Hahn et al., 2005). To facilitate successful reintegration back into the family, natural families are also supported by aftercare groups once the youth returns home (Hahn et al., 2005). Most TFC programs require social workers to carry caseloads of less than 10 families and require a professional mental heath provider to supervise the placement and treatment of children and adolescents receiving TFC services. Social workers also monitor the training and support of the foster parents. TFC personnel may interact not only with the foster family, youth, and natural family, but also with school personnel, probation ofﬁcers, employers, and others in the youth’s community (Hahn et al., 2005). The research base supporting the effectiveness of TFC for youth with SEDs is ambiguous at best (Curtis et al., 2001; Evans et al., 1998a; Hahn et al., 2005; Meadowcroft, Thomlison, & Chamberlain, 1994) and is composed primarily of carefully conducted studies by researchers at the Oregon Social Learning Center (OSLC; Chamberlain & Reid, 1998; Eddy & Chamberlain, 2000; Leve & Chamberlain, 2005). As noted by Farmer et al. (2004), in these studies the results indicated that TFC improved problem behaviors, lowered recidivism rates, and increased the amount of time youth remained in the community for up to 2 years (Chamberlain, Ray, & Moore, 1996; Chamberlain & Reid, 1998; Farmer et al., 2004; Fisher, Gunnar, Chamberlain, & Reid, 2000; Smith, Stormshak, Chamberlain, & Whaley, 2001). In more recent studies of TFC conducted by the OSLC, results continue to indicate positive outcomes including lower deviant peer association, and less antisocial and violent behavior for youth (Eddy & Chamberlain, 2000; Eddy, Whaley, & Chamberlain, 2004; Leve & Chamberlain, 2005a; Smith, 2004). For example, Leve and Chamberlain (2005) found that in comparison to group care, TFC reduced the associations juvenile delinquents had with other delinquent youth. It must be noted, however, that although HCBMHS are designed for youth with SED, the OSLC studies generally treated juvenile delinquents who may or may not have met the two-pronged criteria (functional impairment and symptomatology) for SED (Chamberlain et al., 1996; Chamberlain & Reid, 1998; Eddy et al., 2004; Leve & Chamberlain, 2005a). While

150

TRACY J. PINKARD AND LEONARD BICKMAN

it can be argued that many youth labeled ‘‘juvenile delinquent’’ or being served by the juvenile justice system do have SEDs (Chamberlain & Smith, 2005; Leve & Chamberlain, 2005b; Pullmann et al., 2006), there are also many differences between these two populations. For instance, youth with SED exhibit a wider range of problems (including depression, anxiety, and so forth), whereas those served in the juvenile justice have been reported to primarily suffer from conduct disorder or oppositional deﬁant disorder (Rowland, Halliday-Boykins, & Schoenwald, 2005b). Another issue is whether these studies can be considered effective rather than efﬁcacy studies. While there is not a clear dichotomy, the studies conducted by OSLC have more carefully trained and monitored clinicians and foster parents than typical services of this type (Farmer et al., 2004). Research conducted outside of the OSLC has shown mixed results for TFC (Farmer et al., 2004). In a randomized trial by Evans and colleagues (Evans et al., 1994; Evans et al., 1998a; Evans, Armstrong, Kuppinger, Huz, & McNulty, 1998b), TFC was shown to be less effective than intensive case management in terms of reducing problem behaviors, increasing functioning, and cost. As shown in Table 3 and as Farmer et al. noted, ‘‘currently, the evidence-base for TFC seems to be promising or probably efﬁcacious’’ (2004, p. 862), however not yet well established. Table 3.

Evidence Ratings of the Effectiveness of Home and Community-Based Services. Service

Community-based residential treatment

Pinkard and Bickman (2006)

Promising

Promising

Evaluated but not clear

Evaluated but not clear

Modiﬁed multi-systemic therapy

Promising

Crisis services

Not included

Evaluated but clear Evaluated but clear Evaluated but clear Evaluated but clear Evaluated but clear Promising Promising

Auxiliary and support services

Therapeutic foster care Therapeutic group homes

Farmer et al. (2004)

Family education and support Mentoring Respite care

Coordination and facilitation

Case management Wraparound

Not examined Promising Evaluated but not clear Promising Promising

not not not not not

The Evidence for Home and Community-Based Mental Health Services

151

Therapeutic Group Homes Therapeutic group homes are often not clearly distinguishable from residential treatment centers (RTCs). Whereas RTCs provide services to anywhere from 2 to 100 or more youth, TGHs generally provide services to only 4–10 youth at a time (Burns & Friedman, 1990). Additionally, RTCs may require the youth to be displaced in terms of their community, and TGHs typically are located in or close to the youth’s community allowing the youth to continue attending their local school. RTCs are often visibly distinct from TGH as well; RTCs generally resemble hospitals or other such public facilities, while TGHs are actual houses located in residential neighborhoods (U.S. Public Health Service, 2000). While RTCs rely on several treatment models including a shift staff model in which different staff work in shifts with the youth at the center, TGHs commonly use either the teaching family model developed at the University of Kansas (Farmer et al., 2004) or the Charley model developed at the Menninger Clinic (U.S. Public Health Service, 2000). According to the teaching family model, positive peer interactions, self-government, and trained care staff are thought to lead to improved outcomes. Speciﬁcally, trained couples are employed at the group home 24 hours a day, 7 days a week to provide structured behavioral interventions, teach new skills, and reinforce positive behavior for the youth through individual psychotherapy, group therapy, and behavior modiﬁcation, among other things (U.S. Public Health Service, 2000). Compared with youth being served in TFC, youth served in group homes are generally more likely to be minority males with previous involvement with the juvenile justice system (Curtis et al., 2001). As Farmer et al. (2004) note, there has been scarce research conducted on TGH in the last two decades. Indeed, they and we rate the evidence for the effectiveness of TGHs as ‘‘evaluated but not clear’’ (Farmer et al., 2004, p. 864). Farmer et al. (2004) explain that from the research that has been conducted, ﬁndings suggest that youth being served by TGHs using the teaching family model show increased functioning and decreased symptomatology compared with youth in traditional residential care (Farmer et al., 2004). Drawing from studies comparing TFC to group home settings, TGHs have been found to cost more. In addition, recent studies suggest that in contrast to providing positive peer interaction, group home settings often end up creating more troubled and troubling youth; in the group setting, the youth have little choice but to socialize with other delinquent youth – a known risk factor for delinquency (Leve & Chamberlain, 2005a).

152

TRACY J. PINKARD AND LEONARD BICKMAN

Multi-Systemic Therapy Multi-systemic therapy is based on Bronfenbrenner’s 1979 social-ecological development model and conceptualizes behavior problems in youth as being promoted or diminished by the transactions between the individual, the family, and community factors (Henggeler, Melton, & Smith, 1992). ‘‘Initial therapy sessions are aimed at identifying the strengths and weaknesses of the adolescent, the family, and the family members’ transactions with extrafamilial systems. Identiﬁed problems are explicitly targeted for change, and the strengths of each system are used to facilitate such change.’’ (Henggeler et al.,1991, p. 44)

MST is characterized by short-term treatment lasting 3–5 months (Burns, 2003) that at the individual level entails assessment designed to understand how the identiﬁed problems relate to the youth’s broader context followed by developmentally appropriate interventions intended to target well-deﬁned speciﬁc problems in an action oriented, present-focus manner. At the family level, MST is designed to promote the positive behavior of family members through interventions that require effort by family members on a daily or weekly basis. Additionally, caregivers are empowered to promote family members’ longterm change in terms of the therapeutic intervention. At all levels, MST interventions (1) focus on behaviors that exist within speciﬁc ecological systems as well as between these systems that maintain problem behaviors; (2) are positive and use strengths within the system to promote change; (3) are continuously evaluated; and (4) have an accountability system (Rowland et al., 2005b). In 2002 more than 10,000 in youth in Europe and North America received MST (Henggeler, 2003); this number has substantially increased since that time. This model is one of the ﬁrst programs to be sold commercially and is apparently a ﬁnancial success for the developers. Commercialization may be one of the best ways to insure growth and sustainability. The developers of this program have been pioneers in the self-evaluation of CAMHS programs. There are several studies and reviews that support the effectiveness of MST for serious juvenile offenders (Henggeler, Melton, Brondino, & Scherer, 1997; Henggeler et al., 1986, 1992; Henggeler, Melton, Smith, Schoenwald, & Hanley, 1993; Sutphen, Thyer, & Kurtz, 1995), sexual offenders (Borduin, Henggler, Blaske, & Stein, 1990), suicidal youth (Henggeler, Rowland, Halliday-Boykins, Cunningham, & Pickrel, 2005; Huey et al., 2004), and juvenile offenders with substance use/abuse problems (Borduin et al., 1995; Henggeler et al., 1991; Henggeler, Clingempeel, Brondino, & Pickrel, 2002; Henggeler, Pickrel, & Brondino, 1999a; Schoenwald, Ward, Henggeler, Pickrel, & Patel, 1996).

The Evidence for Home and Community-Based Mental Health Services

153

As traditionally implemented, several modiﬁcations of MST are enacted for use for youth with SED. In particular, with modiﬁed MST, psychiatrists and crisis caseworkers are included in the treatment team rather than consulted as needed (as with traditional MST). Further, access to respite resources is required and there is more frequent supervision than with traditional MST. Caseloads are smaller, treatment adherence is measured more closely, and more extensive training is required than with traditional MST. The effectiveness of modiﬁed MST in treating youth with SEDs has yet to be established. Five studies investigating the use of MST with non-juvenile offenders have been conducted by the originators of MST; four of the studies used the same sample of youth in psychiatric crisis (Henggeler et al., 1999b, 2003; Schoenwald, Ward, Henggeler, & Rowland, 2000). In the ﬁrst study (Henggeler et al., 1999b), 113 youth who were exhibiting symptoms of suicidal ideation, homicidal ideation, or threatened to self-harm or assault others were randomly assigned to either MST (n ¼ 57) or hospitalization (n ¼ 56). Youth were paired (one youth in the MST group, one in the hospital group) in terms of timing of assessments and analyzed using an intent-to-treat approach. Data were collected at three time points – baseline, 2-weeks postrecruitment, and 4-months post-recruitment. Out of 21 possible outcomes, modiﬁed MST was found to be more effective than hospitalization for youth in psychiatric crisis for one-third: caregiver and teacher rated externalizing symptoms, number of days out of school, caregiver rated family cohesion, youth rated family adaptability, and youth and caregiver satisfaction (Henggeler et al., 1999b). For another six outcomes (caregiver self-report of emotional distress, youth reported emotional distress, caregiver and teacher rated internalizing behaviors, caregiver rated social functioning, and caregiver rated family adaptability) both groups improved over time, and for one outcome (self-esteem) the control group improved signiﬁcantly more than the MST group (Henggeler et al., 1999b). In a second study using this sample, Schoenwald, Ward, Henggeler, and Rowland (2000) examined the impact of MST on hospitalization and placement. Findings suggested that at 2-weeks post-referral, 25% of youth in the MST condition required hospitalization and they spent 90% less time hospitalized than youth in the control condition. At four months, this difference in days hospitalized was reduced to 72% less time. In terms of other outof-home placements (OHPs), youth in the MST condition experienced almost half of the days in placement than the youth in the hospitalization condition did and had fewer changes of placement. Schoenwald, et al. (2000) also presented preliminary cost data showing that while hospitalization

154

TRACY J. PINKARD AND LEONARD BICKMAN

costs an average of $6,174 per youth, MST costs slightly less at $5,954 per youth. However, when incremental costs (OHPs) were considered, these ﬁgures changed to $7,878 for hospitalized youth and $8,107 for MST youth. In the third study based on this sample, a 1-year follow-up conducted by (Henggeler et al., 2003), mixed-effects growth curve modeling was used to examine mental health, placement, and caregiver/family relations outcomes for the ﬁnal sample of 156 youth. In terms of internalizing and externalizing symptoms, school attendance, OHPs, and caregiver/family relations, no group differences were found although the change trajectories for the two groups were often different. For example, externalizing symptoms for youth in the MST condition gradually improved, whereas for youth in the hospitalization condition, the change followed a more cubic trajectory with improvement early on followed by a worsening of symptoms then improvement again (Henggeler et al., 2003). In a more recent study using the 115 families in this sample that were receiving Medicaid, Sheidow (2004) evaluated the cost-effectiveness of MST compared with usual services in terms of Medicaid spending. Results indicated that in the short-term signiﬁcantly less was spent on modiﬁed MST, although in the long-term this difference disappeared (Sheidow et al., 2004). Most recently, Rowland et al. (2005) investigated modiﬁed MST with youth in Hawaii. Here 31 youths meeting the criteria for the Felix Decree (a mandate for the provision of mental health services for eligible youth) were randomly assigned to either modiﬁed MST or treatment as usual (TAU). At 6-months, modiﬁed MST was found to be more effective than TAU in 4 out of 14 outcomes: youth rated externalizing and internalizing symptoms, youth reported minor delinquency, and days in OHP (Rowland et al., 2005a). No other statistically signiﬁcant effects were found. However, Littell (2005), in a recent systematic review of studies evaluating MST, concluded that MST did not appear to be more effective than TAU. This review, which appeared in the Campbell Collaboration Systematic Reviews of Interventions and Policy Evaluations, used stringent inclusion criteria including (but not limited to): the program had to be a licensed MST program, participating youth had to be between 10 and 17 years of age, a comparison group had to be used, and there had to be random assignment to treatment groups (Littell, 2005). Of 34 studies identiﬁed in a comprehensive search of electronic databases, reference lists, Internet sites, and personal contacts, only eight studies met all of the inclusion criteria. Based on these eight studies, Littell (2005) concluded ‘‘MST has few if any signiﬁcant effects on measured outcomes, compared with usual services or alternative treatments’’ (p. 457). The results of the meta-analysis were also

The Evidence for Home and Community-Based Mental Health Services

155

published in Children and Youth Services Review and elicited a heated and acrimonious exchange of letters to the editor from both Littell and the MST team (Henggeler, Schoenwald, Swenson, & Borduin, 2006; Littell, 2006). Farmer notes, ‘‘MST probably has the strongest evidence-base among community-based interventions’’ (p. 865). However, further research needs to be conducted with independent evaluators. Given that the most of the research conducted on modiﬁed MST was conducted by the originators of the service on the same sample of youth and the modest results of all of the studies, the evidence of this service for youth with SED should be considered ‘evaluated but not clear’ although Farmer et al. (2004; prior to the Littell publication) rated it as ‘promising’ (see Table 3). Crisis Services Although not included in the Farmer et al. (2004) review, many children and adolescents are referred to mental health services because they are at a point of crisis (U.S. Department of Health and Human Services, 1999), making home and community-based crisis mental health services critical. These crisis services are available 24-hours a day, 7 days a week with the goal of deterring hospitalization or the use of the emergency room by providing immediate short-term care for youth and their families in their home or community. Crisis services are also designed to help the youth and family transition to longer term services if necessary (U.S. Public Health Service, 2000). Generally crisis services are provided for 4–6 weeks at most. While there are several different crisis services available (including walk-in crisis intervention services, home-based crisis intervention (HBCI) services, runaway shelters, telephone hotlines, mobile crisis teams), all crisis services include three basic components: (1) evaluation and assessment, (2) intervention and stabilization, and (3) follow-up planning conducted by staff trained in emergency treatment, family support, and assessment (U.S. Public Health Service, 2000). When used for short-term crisis placements, TGH are considered a crisis service. The evidence-base for crisis services as a whole is built primarily on uncontrolled studies over a decade old (U.S. Department of Health and Human Services, 1999). When only HBCI services are considered, much more recent work has been conducted (Evans et al., 2003). For example, Evans et al. (2003) used a ‘‘randomized multiple treatment group, repeated measures design’’ (p. 94) to compare three crisis programs for 238 youth ages 5–17 years presenting in psychiatric crisis. The ﬁrst program was a HBCI, which includes the three components of crisis services as mentioned

156

TRACY J. PINKARD AND LEONARD BICKMAN

above. This intervention was considered TAU. The second program was enhanced HBCI (HBCI+) that provided cultural competence training for counselors, a family advocate, parent support, respite care, and ﬂexible service money. The third condition was crisis case management. Results indicated that the three programs were roughly equivalent in terms of deterring OHP, and all three groups showed improvement in terms of youth self-concept, family adaptability and cohesion, and parental self-efﬁcacy. No differences were found between the HBCI and HBCI+ groups. However, youth in the crisis case management group showed more signiﬁcant improvement in CBCL internalizing and total scores compared with youth in the other two groups, as well as family cohesion and youth socially supportive behaviors. In ﬁve out of the nine outcomes, no group differences were found. A 2002 study examined a mobile crisis team, short-term residential services, and intensive in-home service. Blumberg (2002) conducted a pre-post study of a crisis intervention program in Delaware. The program, similar to other crisis intervention programs, provided psychiatric evaluations links to other providers, short-term systemic family therapy, and short-term crisis beds for youth when needed. Of the 700 children age 12 or younger involved in the study, none signiﬁcantly harmed themselves or others after starting the program. Over half (n ¼ 465) of the youth were at risk of hospitalization when they started the program. Of these only 2.4% were hospitalized whereas over 90% were referred to outpatient day treatment or other nonrestrictive services. Comparing the ﬁrst 3 years of the crisis intervention program with the 3 years before its implementation, Blumberg found that the mean number of psychiatric beds used decreased by 23%. Additionally, results of cost minimization analyses indicated that the average cost of the hospital diversion aspect of the crisis intervention program (not the program as a whole) was $3,225 per admission compared with the Delaware Division of Child Mental Health Services reported cost of $11,400 per hospital admission. Further, given the reduced number of youth in the crisis intervention program using hospital beds, Blumberg reported that the hospital diversion aspect program saved nearly $20,000 a year (Blumberg, 2002). Several caveats of this study should be mentioned including the obvious fact that no control group was used. The design of the study precludes any statement that the crisis intervention program was the cause of improved outcomes. Additionally, in terms of the cost analyses, costs were only calculated in terms of the hospital diversion aspect of the program, thus the actual costs of the complete program were probably much greater than those reported.

The Evidence for Home and Community-Based Mental Health Services

157

Given the dearth of current research in this area and the methodological shortcomings of the evidence that is available, the evidence rating for crisis services is ‘evaluated but not clear’ (see Table 3). Auxiliary and Support Services Included in the category of home and community-based mental health services for children and adolescents are those services and supports that are provided supplemental to other services (Farmer et al., 2004). Although not designed to speciﬁcally or exclusively effect child outcomes, these services are seen as key aspects to the beneﬁt of home and community-based services in that they provide or increase the support the child receives from the home and the community. Family Education and Support Given the focus of maintaining the child in the home if possible, family education and support services are integral to home and community-based mental health services. The logic underlying these services is that by increasing the family’s knowledge about the youth’s problems and changing the behavior of the family, the child’s behavior will in turn be inﬂuenced (Bickman, Heﬂinger, Northrup, Sonnichsen, & Schilling, 1998; Farmer et al., 2004; Friesen, Pullmann, Koroloff, & Rea, 2005). Based partly on Bandura’s model of self-efﬁcacy, the theory underlying family education and support services suggests, ‘‘Increases in knowledge and self-efﬁcacy should lead, in turn, to greater caregiver involvement in mental health services. It is reasoned that caregiver involvement should also affect the amount of services that the child receives. Because of increased services, it is hypothesized that the mental health status of the child will be improved.’’ (Bickman et al., 1998, p. 271)

The actual components and structure of family education and support services may vary immensely by program. In the Vanderbilt Caregiver Empowerment Project, caregivers participated in an 11-hour training conducted over a 2-week period. It included practice, verbal persuasion, modeling, and vicarious experience along with other training techniques to provide participants with skills, knowledge, and increase their self-efﬁcacy in terms of addressing their youth’s emotional or behavioral problem (s). The training was provided by a team composed of a professional trainer and a parent advocate (Bickman et al., 1998). Although family education and support, according to Burns (2003), has been shown in two randomized clinical trials to increase parental knowledge

158

TRACY J. PINKARD AND LEONARD BICKMAN

and self-efﬁcacy about mental health service use, it is unclear who conducted these studies or how well they were conducted. In Farmer’s 2004 review, only one study was identiﬁed that focused on youth with mental heath problems, the Vanderbilt Caregiver Empowerment Project (Bickman et al., 1998). In a PsycINFOs literature search using the terms ‘‘family education and support’’, ‘‘family empowerment’’, and other related search terms in conjunction with ‘‘mental health’’ and ‘‘child’’, only the Bickman et al. (1998) study was identiﬁed. Using a sub-sample of 250 participants from the Fort Bragg Study, a Department of Defense funded evaluation of the effectiveness of a system of care, Bickman et al. (1998) found that caregivers who received education and support demonstrated increased knowledge and self-efﬁcacy in terms of mental health services compared with those caregivers that did not receive family education and support. However, family education and support had no effect on caregiver involvement in services, increased service use, or improvement in youth mental health status (Bickman et al., 1998). Given the paucity of research conducted in this area, Farmer et al. (2004) rate family education and support as ‘not examined’. Although in terms of respite services, as noted below, Farmer considered one study as evidence of unclear evaluation. However, the Bickman et al. (1998) study did examine the effectiveness of these services in affecting youth outcomes. As such, family education and support is rated as ‘evaluated, but not clear’ (see Table 3). Mentoring Mentoring, also called a recreational service, is an auxiliary support characterized by a relationship between youth and adults that involves regular contact over a period of time (Burns, 2003; DuBois, Holloway, Valentine, & Cooper, 2002; U.S. Department of Health and Human Services, 1999). Mentoring models draw on resilience and social support theories that emphasize the beneﬁt of positive relationships with adults outside of the family (DuBois et al., 2002; Farmer et al., 2004) and assert that ‘‘if caring, concerned adults and role models are available to young people, they will be far more likely to develop into healthy, successful adults themselves’’ (Tierney, Grossman, & Resche, 1995, p. 2). However, the speciﬁc outcomes sought from mentoring programs vary from the general (e.g., promoting positive youth development) to the speciﬁc (e.g., helping youth improve academic outcomes, improve functioning, and so forth; DuBois et al., 2002). Although mentoring programs range in the outcomes they seek, most, if not all, programs require prospective mentors to undergo background checks

The Evidence for Home and Community-Based Mental Health Services

159

and screening interviews, and many require ongoing training and supervision of mentors. Currently there are nearly 2,000 organizations that offer mentoring programs for children and adolescents (Farmer et al., 2004) with Big Brothers/Big Sisters of America being the most well known. Similar to the research conducted evaluating family education and support services, the evidence-base for mentoring is built on studies that have examined several at-risk populations rather than youth with SED exclusively. In a meta-analysis of 55 evaluations of youth mentoring programs, Dubois et al. (2002) reported that participation in mentoring programs produced moderate to small positive effects for youth in terms of problem/ high-risk behavior, emotional/psychological adjustment, social competence, academic outcomes, and employment (average effect size of 0.18 in a random-effects model and 0.14 in a ﬁxed effects model). However, for youth experiencing only individual risk factors (e.g., psychosocial impairment) as opposed to either environmental risk factors (e.g., single parent home) or both, mentoring had virtually no effect: ‘‘Finally, near-zero average estimates of effect size were evident for those samples of youth indicated to be experiencing individual risk factors alone (effect sizes of .00 and .03 [for the ﬁxed-effects model and random-effects model respectively])’’ (DuBois et al., 2002, p. 182). Further analyses suggest that for youth with individual risk factors participating in poorly implemented mentoring programs the effects were actually negative (average effect size of 0.06 in the random-effects model and 0.13 in the ﬁxed effects model). Additionally, it must be noted that of the programs reviewed, only 11 included follow-up assessments. Thus it is unclear whether the modest effects found for mentoring in general are sustained over time. The Farmer et al. (2004) review cites the evidencebase for mentoring as ‘‘promising’’. However given the potential for negative effects found in the DuBois et al. (2002) review, the evidence-base rating is ‘evaluated, but unclear’ as shown in Table 3. Respite Services Although often thought of as a service for caregivers, respite care is recognized as a home and community-based mental health service. Respite is also designed to beneﬁt the youth by providing positive social experiences and other siblings as well by offering them more time to interact with caregivers (Bruns & Burchard, 2000). There are two types of respite care: inhome and out-of-home. The difference being that in in-home models respite is provided by a trained respite worker who comes into the home on a regular basis, while out-of-home respite entails the respite worker taking the youth into his or her home or the community for a speciﬁed amount of time

160

TRACY J. PINKARD AND LEONARD BICKMAN

(Bruns & Burchard, 2000). Out-of-home respite may or may not include overnight stay (Boothroyd, Kuppinger, Evans, Armstrong, & Radigan, 1998). As noted by Bruns and Burchard (2000) ‘‘research on the effectiveness of respite care continues to be sparse, with the majority of studies that do report outcomes suffering from serious methodological deﬁciencies (p. 40)’’. Generally, the research that has been conducted has been on populations other than youth with SED and has found respite care to result in reduced family stress (Burns, 2003) but to have little or no effect on child outcomes (Friesen et al., 2005). To date only one study, a quasi-experimental study by Bruns and Burchard (2000), has examined the impact of stand-alone respite care (in contrast to respite care provided in conjunction with or as a component of other services). In this study 73 youth were assigned to either respite care (n ¼ 45) or a wait-list control group (n ¼ 28). Results indicated that in terms of OHP, signiﬁcantly fewer youth in the respite group required OHP than in the control group. Additionally, youth in the respite group used signiﬁcantly fewer OHP days and were rated by their caregivers as likely to need future OHP (Bruns & Burchard, 2000). The respite group also showed signiﬁcant improvement on one (community externalizing behaviors) of the four clinical outcomes and one of the seven caregiver strain outcomes compared with the control group. There were no group differences in the number of youth using crisis services, or family functioning, the remaining clinical outcomes, or the remaining six caregiver strain outcomes (Bruns & Burchard, 2000). Given the scant research base and modest ﬁndings the evidence-base for respite services we categorized it as ‘evaluated but not clear’ (see Table 3) in accord with Farmer et al. (2004). Coordination and Facilitation Also called operational services, home and community-based mental health services that fall into this category are those that emphasize providing access to and coordinating services for youth (Farmer et al., 2004). These services are key components of Systems of Care (Stroul & Friedman, 1994). Included are intensive case management (ICM) and wraparound. Intensive Case Management In 1986, Congress began to allow states to use Medicaid funds to cover ‘‘targeted case management’’ services. These services were to be freestanding of other services and to be provided to speciﬁc groups such as children and

The Evidence for Home and Community-Based Mental Health Services

161

adolescents with SED, homeless adults, and so forth. Often used in conjunction with other services, ICM mobilizes, links, and sustains services and support for youth. Although there are several models of case management including ‘‘the generalist as service broker model, primary therapist, interdisciplinary team, comprehensive service center, family as case manager, supportive care, and volunteer as case manager’’ (Farmer et al., 2004, p. 866), some of the more salient characteristics of ICM include small case loads, no treatment time limits, and ﬂexible funds (Burns, 2003; Cauce, Morgan, Wagner, & Moore, 1994; Evans, Boothroyd, & Armstrong, 1997). Additionally, regardless of delivery model, most case managers provide advocacy, assessments, planning, service linkages, and so forth, to families (Evans et al., 1994, 1998a, 1998b). According to Burns (2003), the evidence-base for ICM is supported by four randomized trials and three quasi-experimental studies that show ICM to result in less restrictive placements and some increased functioning. The majority of the studies were conducted by Evans and colleagues on Family Centered Intensive Case Management in New York and suggest that youth who received ICM accessed a greater array of services and more often than control group youth and at lower costs (Evans et al., 1994, 1997, 1998a, 1998b, 2003; Evans, Armstrong, & Kuppinger, 1996). In a randomized control trial comparing 27 youth receiving ICM and 15 youth receiving TFC, Evans et al. (1998) found that children and adolescents in the ICM group experienced improved functioning and fewer problem behaviors compared with those in the control group and at a lower cost. Additionally, in a study comparing two home-based crisis intervention services to crisis case management services, for four out of 9 outcomes, youth in crisis case management showed signiﬁcant improvement compared with youth in either of the other two groups while there were no signiﬁcant differences between the groups on the other ﬁve outcomes (Evans et al., 2003). However, Farmer et al. (2004) note that outside of the small-scale research by this one group, there is little evidence supporting the effectiveness of ICM compared with TAU. Additionally, in these studies, intervention ﬁdelity is not extensively documented so it is unclear exactly which model of case management is being evaluated: ‘‘One of the challenges of making sense of the literature on case management is understanding the model of case management which has been used as the intervention’’ (Evans et al., 1998a, p. 9). As shown in Table 3, the evidence-base for intensive case management is promising, and Farmer et al. (2004) agree.

162

TRACY J. PINKARD AND LEONARD BICKMAN

Wraparound Wraparound is often viewed more as a philosophy than as a particular service (Bickman, Smith, Lambert, & Andrade, 2003; Burns, Schoenwald, Burchard, Faw, & Santos, 2000; Burns & Goldman, 1999; Myaard, Crawford, Jackson, & Alessi, 2000) and is characterized by ‘‘a deﬁnable planning process that involves the child and family. It is designed to provide a unique set of individualized community services and natural supports based on a set of core principles and values’’ (Farmer et al., 2004, p. 867). Developed in the mid-1980s, some researchers consider wraparound and ICM as one and the same (Burns, 2003). Focused on youth with emotional and behavioral problems that place them at-risk for OHP, the wraparound philosophy is based on the theory of environmental ecology (Munger, 1998). It holds that youth function best when their immediate family environment is coordinated with the larger service system (Burns et al., 2000). Wraparound is thought to be comprised of ten principles that generally promote culturally competent, individualized care for youth that are based on their natural supports (especially family) and situated in their community. Unlike many of the other home and community-based mental health services for children and adolescents, there is no deﬁned intervention duration for wraparound. Most of the studies comprising the evidence-base for wraparound are unpublished, only pre-post design or both. There also is not a consensus on whether there is a conceptual difference between ICM and wraparound. Some researchers consider wraparound and case management synonymous (Burns et al., 2000; Clark, Lee, Prange, & McDonald, 1996; Clark, Prange, Lee, McDonald, & Boyd, 1998; Evans et al., 1996, 1998b). Instead of making a distinction between the two, they include studies of the effectiveness of case management in the evidence-base for wraparound. For example, Burns (2003) cited studies by Evans and colleagues (Burns et al., 2000; Clark et al., 1996, 1998; Evans et al., 1996, 1998b) in her review of wraparound and MST, whereas in these discussions, those studies are included in the section on case management above. Of the randomized or quasi-experimental studies, ﬁndings have generally indicated that wraparound is no more or less effective than TAU (Bickman et al., 2003; Clark et al., 1996, 1998). In a study by Clark et al. (1996, 1998) comparing wraparound enhanced foster care with regular foster care for children at risk for or experiencing behavioral or emotional problems, youth were randomly assigned to groups. Results indicated that wraparound enhanced foster care was effective in improving some outcomes for some subsets of youth. Speciﬁcally, youth in the wraparound group experienced fewer

The Evidence for Home and Community-Based Mental Health Services

163

runaway days and days incarcerated than youth in the control group. Additionally, older youth in the wraparound condition were signiﬁcantly more likely to have a permanent-type ﬁnal placement (e.g., with parents or family as opposed to an emergency shelter) and more likely to spend more time in such a setting compared with older youth in the control group. Males in the wraparound group were also found to show signiﬁcant improvement in terms of conduct disorder and broadband externalizing problems. However, females in the control group faired better than intervention group females in terms of these two outcomes (conduct disorder and broadband externalizing problems). For many (11 out of 19) measured outcomes, no signiﬁcant group differences were found (Clark et al., 1996, 1998). In one of the few wraparound evaluations not done by an originator of the philosophy, Bickman et al. (2003) conducted a Department of Defense funded quasi-experimental evaluation of a managed care company’s wraparound demonstration for youth ages 4–16. Youth receiving wraparound were found to receive more case management, in-home and non-traditional services than youth receiving TAU. However, in terms of youth symptoms, functioning, well-being, and several other outcomes, no statistically signiﬁcant differences were found between the groups. Furthermore, ‘‘on average, service costs for children in the TAU comparison group were 42% less than those for a typical child in the Demonstration’’ (p. 149). This study was hampered by the lack of random assignment and its inability to document the presence of all the wraparound characteristics thought be important. As reported in Table 3 and by Farmer et al. (2004), the present evidence for wraparound is ‘‘on the weak side of promising.’’ It is worth noting that unlike most of the other HCBMHS, there are measures available to assess adherence to the wraparound service model. Both the Wraparound Observation Form (Epstein et al., 1998) and the Wraparound Fidelity Index (Bruns, Suter, Burchard, Force, & Leverentz-Brady, 2004) measure implementation of wraparound services. As mentioned above however, while these measures allow us to document ﬁdelity to the service model, they do not offer insight into the speciﬁc treatment delivered.

HALF EMPTY? HALF FULL? Lack of Clarity of Terms This chapter has focused on the effectiveness of home and community-based services. From the glass half full perspective several approaches to providing

164

TRACY J. PINKARD AND LEONARD BICKMAN

services in home and community settings are promising and do not appear to cause harm. Moreover, families, youth, and providers appear satisﬁed with these services. Many, who advocate for these services, while recognizing some of the methodological problems, consider the scientiﬁc evidence the best available and thus sufﬁcient. Critics of this perspective opine that the best evidence available is not a sufﬁcient criterion for the continued expansion of these services. The research must meet not relative criteria but absolute quality criteria to be taken seriously. Most of the studies of HBCBS use a two-point pre-post design that makes it impossible to conﬁdently determine causality. Moreover, the use of only two time points makes the interpretation of change ambiguous. If the glass is seen as half empty, the studies have failed to demonstrate that these services are effective in producing better outcomes for youth. We believe that there is currently little solid evidence of the effectiveness of any home and community-based mental health service for children and adolescents with SED. None of the services received a rating of ‘well-established’ or ‘better/best treatment’. While advocates of these services point out evidence showing that many of the services increase access and service satisfaction, these ﬁndings do not generalize to clinical outcomes such as reduction in symptoms or an increase in functioning. Moreover, positive ﬁndings are tempered by critics who stress that the majority of the studies forming the evidence-base for HCBMHS for youth have critical methodological problems and have not been independently evaluated. As such, most of these services at best are rated ‘‘evaluated but not clear’’. In addition, very little distinction has been made between services and treatments; it is unclear whether the services are effective, or whether the treatments provided in the services are effective. If all services used manualized treatments then the ﬁdelity and quality of the treatment could be examined. Since this is not the case, it is hard to determine how to monitor services without undue burden to the clinician or client. The procedures used in efﬁcacy studies, such as videotaping, are not feasible. Without at least a theory guiding treatment, let alone a manual, it is extremely difﬁcult to know what to assess to determine ﬁdelity or quality. In addition the conﬂation of the terms ‘‘services’’ and ‘‘treatments’’ has led to widespread confusion. A more reﬁned set of distinctions could be useful for program planning and policy. For example, distinctions can be made between services as settings (i.e., school-based mental health, home-based services, TGH); program services that are treatments, i.e. manualized, codiﬁed, capable of being used for training (i.e., some family education and support models, the

The Evidence for Home and Community-Based Mental Health Services

165

Chamberlain TFC model and some speciﬁc Cognitive Behavioral Therapies or medication therapies). We do not think that the general philosophy or orientation of the clinician, such as psychodynamic or cognitive behavioral provides sufﬁcient information to describe the treatment provide. There also needs to be more clarity in describing who provides the service or treatment. The literature uses the terms counselor, therapist, clinician, and provider interchangeably. Since not all persons who provide mental health treatment are the same we need some systematic way of describing them. Should We Continue Current Policies? As Hoagwood et al. (2001) noted, ‘‘the use of the term ‘evidence-based practice’ presupposes agreement as to how the evidence was generated, what the evidence means, and how or when the practice can be implemented’’ (p. 1179). These issues raise several questions: What action should be taken when the evidence of the effectiveness of a current service is not established? Should service provision be suspended until ‘‘adequate’’ evidence is available? Should services be widely disseminated and funded despite the lack of a strong evidence-base supporting their effectiveness? Based on current events the answer seems to be to continue providing and expanding these services regardless of the strength of the evidence. One argument for continuing services regardless of their effectiveness is that they may be better than no services at all. Why change services that are seen as acceptable, especially if they are a good ﬁt with the values of leaders? Even without strong supporting evidence it is not reasonable to argue for stopping current services especially without an alternative. However, the cost of these services must also be considered, as more than $11 billion is spent annually on mental health services for youth (Ringel & Sturm, 2001). Are we spending these limited resources in way that maximizes the beneﬁts to youth and their families? What is the opportunity cost of following our current practice? The costs of providing many of these services in the context of a Systems of Care has been found to be greater than the cost of traditional care (Foster & Connor, 2005). Although the Systems of Care movement purports that coordinating and linking multiple services would result in increased service access and improved outcomes for youth (Stroul & Friedman, 1986), the Ft. Bragg and Stark County studies conducted by Bickman and his colleagues found that although providing a Systems of Care did increase access to services, the clinical outcomes for youth in the system of care had no

166

TRACY J. PINKARD AND LEONARD BICKMAN

better outcomes than those receiving treatment as usual. Moreover, the system of care was signiﬁcantly more expensive (Bickman, 1996a, 1996b; Bickman et al., 1994, 1995). At a 5-year follow-up (Bickman, Lambert, Andrade, & Penzaloza, 2000), there were still no signiﬁcant differences in outcomes between youth treated in the Systems of Care and those treated in uncoordinated services. Despite the results of these evaluations, coordination of services is still cited as one of the ten values of children’s mental health (Huang et al., 2005). In a recent class action lawsuit in Massachusetts, the court, in explaining its decision to support state-wide dissemination of HCBMHS and systems of care stated that ‘‘prompt, coordinated services that support a child’s continuation in the home can allow even the most disabled child a reasonable chance at a happy, fulﬁlling life’’ (Rosie et al. v. Mitt Romney et al., 2006, p. 8). An Example of a Bad Investment Funding and expansion of these services is continuing regardless of the absence of clear scientiﬁcally valid support. The most egregious example of this is The Comprehensive Community Mental Health Services Program for Children and Their Families that is funded by SAMHSA. There are currently 85 grant communities and 31 former grant programs. The cost of this program from 1995 to 2006 is over one billion dollars (yes billion), not including the cost of government personnel. We were able to obtain this information from SAMHSA under the Freedom of Information Act (SAMHSA, 2005). This is the single most expensive mental health program for children and adolescents ever implemented. But after more than a decade of operation what do we know about its effectiveness? Has this be a wise investment of funds, especially in an area that is grossly underfunded? Answering these questions has not been easy. The evaluation contractor from the beginning of this program, ORC Macro, is required to submit annual reports to Congress. The latest report available (Center for Mental Health Services, 2001) describes a longitudinal comparison study of three ‘‘matched’’ communities. The study was designed to answer the question whether children treated in a system of care have better mental health outcomes than children receiving care as usual. The results of only one of the comparisons (Stark County) were described. Of the seven major measures used only one subscale showed a difference favoring the system of care. In general, the pattern of results was similar to the earlier randomized experimental study conducted in Stark

The Evidence for Home and Community-Based Mental Health Services

167

County by Bickman and his colleagues. Both the system of care and comparison children improved at about the same rate. Moreover, the system of care was much more expensive. This was also the case for a similar ORC Macro Phase II comparison study of two additional communities. Although this evaluation was supposed to have been completed several years ago we are not aware of a single scientiﬁcally credible outcome evaluation published in a peer-reviewed journal. The quality and quantity of the evidence supporting the System of Care program does not justify an expenditure of over a billion dollars. Balancing Values The critical difference between the half empty and half full perspectives may be in how to balance values. Support for home and community-based services appears to be based on congruence with such values as cultural competence, parent involvement and placement in the least restrictive environment. Both systems of care and wraparound have been characterized as value systems or philosophies by their proponents (Burns, 2003; Stroul & Friedman, 1994). From this perspective, the design and the results of evaluations of effectiveness are not as relevant since it is not better outcomes that appear to support the continued existence and expansion of these services but congruence with these values. Parental involvement in care is considered a critical component of these services. The decision to expend resources on parental involvement is not based on scientiﬁc data but on a value system that speciﬁes that it is important to include parents in treatment. This perspective may judge client outcomes as a lower priority than congruence with other values. It might also explain the continued support for the expansion of these services in the absence of scientiﬁc evidence of effectiveness. In contrast, the half empty perspective places a higher value on the scientiﬁc evidence. In this perspective, the ﬁndings of studies are greatly qualiﬁed by the soundness of the methodology and analysis. This perspective sees scientiﬁc values as paramount. Thus, judgment about the expansion of services is dependent on the values of the perceiver, making it extremely difﬁcult to decide if there is a right or wrong answer to the question of continued support of HCBMHS. Should a greater emphasis be placed on the scientiﬁc evidence than congruence with other values such as those articulated in Systems of Care and wraparound philosophies? The claims from both perspectives are severely limited given the number and quality of research in this area.

168

TRACY J. PINKARD AND LEONARD BICKMAN

OTHER GLASSES As Weisz, Sandler, Durlak, and Anton (2005) indicate, the limitations of the evidence-base suggest possible future directions for the ﬁeld. One such direction is the development of new mental health services for children and adolescents. However, the lag between innovation and implementation is great (Huang et al., 2005; Lambert, Hansen, & Finch, 2001b; Weisz et al., 2005). Weisz et al. (2005) explain ‘‘many years must pass for an intervention idea to become a program, for the program to be tested, and for the test to be published; the cycle must be repeated for a body of evidence to build, and many more years will pass before signiﬁcant dissemination can be organized.’’ (p. 642)

Client-Focused Research Rather than to continue a fruitless debate over whether the glass is half full or half empty, or whether the current evidence-base for the effectiveness of HCMBHS is sufﬁcient to support their continued use and dissemination, some researchers and practitioners are turning toward patient-focused research as another approach to providing mental health services (Lambert et al., 2001b, 2001c). Although clinical trials research (efﬁcacy research) emphasizes the average response of patients to treatment in highly controlled experimental conditions, and effectiveness research focuses on the mean response of patients in naturalistic settings, client or patient-focused research attempts to answer the question, ‘‘Is this particular treatment working for this patient?’’ (Lambert et al., 2001b, p. 159). Patient-focused research is supported by a continuous quality improvement (CQI) approach (Bickman, 1999; Bickman & Noser, 1999; Huang et al., 2005; Lambert et al., 2001b; Schoenwald, Sheidow, & Letourneau, 2004). Adopting a CQI strategy entails ‘‘assessment, feedback, and application of information’’ (Bickman, 1999, p. 973) on a global level and on a more speciﬁc level would require that ‘‘clinicians (a) know outcomes, (b) receive feedback, (c) know their own treatment strategies, (d) connect process with outcome, (e) ﬁt knowledge to individuals, (f) generalize, and (g) apply their knowledge’’ (Bickman, 1999, p. 969). CQI also requires that the approach be adopted at an organizational level so that it can be nurtured, valued, and sustained.

The Evidence for Home and Community-Based Mental Health Services

169

Assessment and Outcomes and Clinical Processes The ﬁrst element of patient-focused research, knowing outcomes, is based on reliable and repeated assessment (Lambert et al., 2001a, 2001b). Lambert, Ogles, and Masters (1992) state that ‘‘patterns of client change and the associated effectiveness of our interventions y must be measured and quantiﬁed in a manner that will let us make clear statements regarding the type and magnitude of change experienced by our clientele’’ (p. 527). Key to strengthening service accountability and improving service quality is ongoing data collection based on a reliable measurement system (Huang et al., 2005). We also believe that psychopathology outcome data is not sufﬁcient to manage optimally clinical treatment. Feedback should also be provided on strengthbased outcomes such as hopefulness as well as processes that are thought to be fundamental to success such as therapeutic alliance. Information about the processes and progress of the client should be concurrent with treatment. Feedback A second key element of patient-focused research is feedback to the clinician and supervisors (Huang et al., 2005; Lambert et al., 2001c). Unfortunately, most practitioners are currently trained, supervised, and practice without objective information about client treatment progress (Bickman, 1999; Bickman & Noser, 1999; Huang et al., 2005; Lambert et al., 2001b; Schoenwald et al., 2004). Feedback about client progress that is based on systematic assessment may greatly improve client outcomes by helping clinicians learn from experience in the long-term, and alerting them to problem patients in the short-term (Bickman, 1999; Bickman & Noser, 1999; Huang et al., 2005; Lambert et al., 2001b; Schoenwald et al., 2004). Lambert et al. (2001) tested the impact of color-coded feedback to therapists on psychotherapy outcomes in a randomized sample of 609 university students. Patients who were not improving at the expected rate or level of progress were called ‘signal cases’ and were indicated by yellow and red dots on feedback graphs; patients who were improving as expected were indicated with white or green dots. Signal cases with therapists that received feedback improved at higher rates (26%) and deteriorated at signiﬁcantly lower rates (6%) than signal cases with therapists that did not receive feedback. Additionally, feedback was found to inﬂuence treatment duration. Signal cases in the feedback group attended signiﬁcantly more sessions (9.68) than those in any other group; patients who were progressing as

170

TRACY J. PINKARD AND LEONARD BICKMAN

expected in the feedback group attended signiﬁcantly fewer sessions (2.81) than any other group. In other words, feedback seems to increase the efﬁciency of treatment – when therapists received feedback those patients that needed more sessions received them, and those who did not need longer treatment duration did not receive it. Application of Information The third element of patient-focused research is application of the information supplied by the assessment and the feedback. This process includes (as listed above) knowing treatment strategies, connecting process with outcome, ﬁtting this knowledge to the individual client, and generalizing when needed (Bickman, 1999). Knowledge of evidence-based treatment strategies allows appropriate application of the information provided by the assessment and feedback. Application also requires ﬁtting this knowledge with the speciﬁc client and connecting the therapy process with the outcomes. Finally, application of the information obtained through assessment and feedback entails generalizing to other situations and clients when necessary. As Silverman, Kurtines, and Hoagwood (2004) explain, ‘‘Having an understanding of what needs to be changed and how to change it provides the clinician in practice with the ﬂexibility to variation in problem, context, or condition’’ (p. 296). Contextualized Feedback Intervention and Training (CFIT) CFIT is a working example of a patient-focused CQI solution to the evidence challenge (Bickman, Riemer, Breda, & Kelley, in press; Riemer, Rosof-Williams, & Bickman, 2005; Sapyta, Riemer, & Bickman, 2005). In congruence with CQI philosophy that emphasizes ‘‘understanding and improving the underlying work processes and systems rather than on correction of individuals’ mistakes after the fact’’ (Schoenwald et al., 2004, p. 94), the CFIT system entails a complete loop of assessment, interpretation, solution, and feedback in a timely manner. Clinical practice should be enhanced through feedback in conjunction with access to online training that focuses on affecting factors common to all therapy modalities. Currently, we are conducting a large-scale evaluation of the CFIT system in over 20 states and 40 sites. CFIT addresses practitioners, organizations, and researchers; hopefully this alternative approach to treatment will be successful. It is unlikely that additional studies of vaguely conceptualized HCBMHS are going to be productive as long as we are not more precise in deﬁning

The Evidence for Home and Community-Based Mental Health Services

171

these services. CBMHS is a merely a setting and it is unlikely that setting qua setting is going to improve outcomes. There is deﬁnitely a need for more studies of some of these service treatments – in particular TFC for youth with SED; modiﬁed MST for youth with SED; family education and support; multi-family groups; crisis management programs, etc. Given the current political reality it may not be possible slow down the expansion of HCBMHS. States are increasingly reliant upon Medicaid for funding outpatient services. Insofar as states continue to rely heavily on Medicaid funding, the expansion of HCBMHS will continue. A more productive strategy is to try to compete with these services by developing effective alternatives. The researchers in this ﬁeld are not lacking in creativity. There are some promising approaches such an emphasis on common factors (Karver et al., 2005), identiﬁcation of effective components of evidence-based treatments (Chorpita, Daleiden, & Weisz, 2005) and the continued effort to apply evidence-based treatments developed in the laboratory (Jensen et al., 2005).

ACKNOWLEDGMENT This research was supported by training grant T32 MH019544-13 from the National Institute of Mental Health to Leonard Bickman.

REFERENCES Bickman, L. (1996a). A continuum of care: More is not always better. American Psychologist, 51(7), 689–701. Bickman, L. (1996b). Implications of a children’s mental health managed care demonstration evaluation. Journal of Mental Health Administration, 23(1), 107–117. Bickman, L. (1999). Practice makes perfect and other myths about mental health services. American Psychologist, 54(11), 963–978. Bickman, L., Guthrie, P. R., Foster, E. M., Lambert, E. W., Summerfelt, W. T., Breda, C., & Heﬂinger, C. (1994). Final report of the outcome and cost/utilization studies of the Fort Bragg evaluation project: Volume 1. Nashville, TN: Vanderbilt Institute for Public Policy Studies, Center for Mental Health Policy. Bickman, L., Guthrie, P. R., Foster, E. M., Lambert, E. W., Summerfelt, W. T., & Heﬂinger, C. A. (1995). Evaluating managed mental health services: The Fort Bragg experiment. New York: Plenum Press. Bickman, L., Heﬂinger, C., Northrup, D. A., Sonnichsen, S. E., & Schilling, S. (1998). Long term outcomes to family caregiver empowerment. Journal of Child and Family Studies, 7, 269–282.

172

TRACY J. PINKARD AND LEONARD BICKMAN

Bickman, L., & Noser, K. (1999). Meeting the challenges in the delivery of child and adolescent mental health services in the next millennium: The continuous quality improvement approach. Applied and Preventive Psychology, 8, 247–255. Bickman, L., Riemer, M., Breda, C., & Kelley, S. D. (in press). CFIT: A system to provide a continuous quality improvement infrastructure through organizational responsiveness, measurement, training, and feedback. Report on Emotional & Behavioral Disorders in Youth. Bickman, L., Smith, C. M., Lambert, E., & Andrade, A. R. (2003). Evaluation of a Congressionally mandated wraparound demonstration. Journal of Child and Family Studies, 12(2), 135–156. Blumberg, S. (2002). Crisis intervention program: An alternative to inpatient psychiatric treatment for children. Mental Health Services Research, 4(1), 1–6. Boothroyd, R. A., Kuppinger, A. D., Evans, M. E., Armstrong, M. I., & Radigan, M. (1998). Understanding respite care use by families of children receiving short-term, in-home psychiatric emergency services. Journal of Child and Family Studies, 7(3), 353–376. Borduin, C. M., Henggler, S. W., Blaske, D. M., & Stein, R. J. (1990). Multisystemic treatment of adolescent sexual offenders. International Journal of Offender Therapy and Comparative Criminology, 35, 105–114. Borduin, C. M., Mann, B. J., Cone, L. T., Henggeler, S. W., Fucci, B. R., Blaske, D. M., et al. (1995). Multisystemic treatment of serious juvenile offenders: Long-term prevention of criminality and violence. Journal of Consulting and Clinical Psychology, 63(4), 569–578. Bruns, E., Suter, J. C., Burchard, J. D., Force, M., & Leverentz-Brady, K. (2004). Assessing ﬁdelity to a community-based treatment for youth: The Wraparound Fidelity Index. Journal of Emotional and Behavioral Disorders, 12, 69–79. Burns, B., Schoenwald, S. K., Burchard, J., Faw, L., & Santos, A. B. (2000). Comprehensive community-based interventions for youth with severe emotional disorders: Multisystemic therapy and the Wraparound Process. Journal of Child and Family Studies, 9(3), 283–314. Burns, B. J. (1991). Mental health service use by adolescents in the 1970s and 1980s. Journal of the American Academy of Child and Adolescent Psychiatry, 30(1), 144–150. Burns, B. J. (2003). Children and evidence-based practice. Psychiatric Clinics of North America, 26(4), 955–970. Burns, B. J., & Friedman, R. M. (1990). Examining the research base for child mental health services and policy. The Journal of Mental Health Administration, 17(1), 87–98. Burns, B. J., & Goldman, S. K. (1999). Promising practices in wraparound for children with serious emotional disturbances and their families (Vol. IV). Washington, DC: Center for Effective Collaboration and Practice, American Institutes for Research. Bruns, E., & Burchard, J. (2000). Impact of respite care services for families with children experiencing emotional and behavioral problems. Children’s Services: Social Policy, Research, and Practice, 3(1), 39–61. Cauce, A. M., Morgan, C. J., Wagner, V., Moore, E., Wurzbacher, K., Weeden, K., Tomlin, S., & Blanchard, T. (1994). Effectiveness of intensive case management for homeless adolescents: Results of a 3-month follow-up. Journal of Emotional and Behavioral Disorders, 2(4), 219–227. Center for Mental Health Services. (2001). Annual report to congress on the evaluation of the comprehensive community mental health services for children and their families program, 2001. Atlanta, GA: ORC Macro.

The Evidence for Home and Community-Based Mental Health Services

173

Centers for Medicare and Medicaid Services. (2004). Home and community-based services waiver program. From www.cms.hhs.gov/medicaid/1915c/history.asp Centers for Medicare and Medicaid Services. (2006). HCBS Waivers Section 1915 (c). From http://www.cms.hhs.gov/MedicaidStWaivProgDemoPGI/05_HCBSWaivers-Section1915(c) Chamberlain, P., Ray, J., & Moore, K. J. (1996). Characteristics of residential care for adolescent offenders: A comparison of assumptions and practices in two models. Journal of Child and Family Studies, 5(3), 285–297. Chamberlain, P., & Reid, J. B. (1998). Comparison of two community alternatives to incarceration for chronic juvenile offenders. Journal of Consulting and Clinical Psychology, 66(4), 624–633. Chamberlain, P., & Smith, D. K. (2005). Multidimensional treatment foster care: A community solution for boys and girls referred from juvenile justice. In: E. D. Hibbs & P. S. Jensen (Eds), Psychosocial treatments for child and adolescent disorders: Empirically based strategies for clinical practice (pp. 557–573). Washington, DC: American Psychological Association. Chambless, D., & Hollon, S. D. (1998). Deﬁning empirically supported therapies. Journal of Consulting and Clinical Psychology, 66(1), 7–18. Chambless, D., Sanderson, W. C., Shoham, V., Bennett Johnson, S., Pope, K. S., Crits-Christoph, P., Baker, M., Johnson, B., Woody, S. R., Sue, S., Beutler, L. E., Williams, D. A., & McCurry, S. (1996). An update on empirically validated therapies. The Clinical Psychologist, 49(2), 5–18. Clark, H. B., Lee, B., Prange, M. E., & McDonald, B. A. (1996). Children lost within the foster care system: Can wraparound service strategies improve placement outcomes. Journal of Child and Family Studies, 5(1), 39–54. Clark, H. B., Prange, M. E., Lee, B., McDonald, B. A., & Boyd, L. A. (1998). An individualized wraparound process for children in foster care with emotional/behavioral disturbances: Follow-up ﬁndings and implications from a controlled study. In: M. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with behavioral and emotional disorders and their families: Programs and evaluation best practices (pp. 513–542). Austin, TX: Pro-Ed Publishing. Coalition for Evidence-Based Policy. (2003). Identifying and implementing educational practices supported by rigorous evidence: A user friendly guide. Washington, DC: US Department of Education Institute of Education Sciences National Center for Education Evaluation and Regional Assistance. Curtis, P. A., Alexander, G., & Lunghofer, L. A. (2001). A literature review comparing the outcomes of residential group care and therapeutic foster care. Child and Adolescent Social Work Journal, 18(5), 377–392. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30(2), 157–197. Eddy, J. M., & Chamberlain, P. (2000). Family management and deviant peer association as mediators of the impact of treatment condition on youth antisocial behavior. Journal of Consulting and Clinical Psychology, 68(5), 857–863. Eddy, J. M., Whaley, R. B., & Chamberlain, P. (2004). The prevention of violent behavior by chronic and serious male juvenile offenders: A 2-year follow-up of a randomized clinical trial. Journal of Emotional and Behavioral Disorders, 12(1), 2–8.

174

TRACY J. PINKARD AND LEONARD BICKMAN

Epstein, M. H., Jayanthi, M., McKelvey, J., Frankenberry, E., Hardy, R., Dennis, K., & Dennis, K. (1998). Reliability of the wraparound observation form: An instrument to measure the wraparound process. Journal of Child and Family Studies, 7(2), 161–170. Evans, M. E., Armstrong, M. I., Dollard, N., Kuppinger, A. D., Huz, S., & Wood, V. M. (1994). Development and evaluation of treatment foster care and family-centered intensive case management in New York. Journal of Emotional and Behavioral Disorders, 2(4), 228–238. Evans, M. E., Armstrong, M. I., & Kuppinger, A. D. (1996). Family-centered intensive case management: A step toward understanding individualized care. Journal of Child and Family Studies, 5(1), 55–65. Evans, M. E., Armstrong, M. I., Kuppinger, A. D., Huz, S., & Johnson, S. (1998a). A randomized trial of family-centered intensive case management and family based treatment: Final report. Tampa, FL: University of South Florida. Evans, M. E., Armstrong, M. I., Kuppinger, A. D., Huz, S., & McNulty, T. L. (1998b). Preliminary outcomes of an experimental study comparing treatment foster care and family-centered intensive case management. In: M. H. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with behavioral and emotional problems and their families: programs and evaluation best practices (pp. 543–580). Austin, TX: Pro-Ed Publishing. Evans, M. E., Boothroyd, R. A., & Armstrong, M. I. (1997). Development and implementation of an experimental study of the effectiveness of intensive in-home crisis services for children and their families. Journal of Emotional and Behavioral Disorders, 5(2), 93–105. Evans, M. E., Boothroyd, R. A., Armstrong, M. I., Greenbaum, P. E., Brown, P. C., & Kuppinger, A. D. (2003). An experimental study of the effectiveness of intensive in-home crisis services for children and their families: Program outcomes. Journal of Emotional and Behavioral Disorders, 11(2), 93–104. Farmer, E. M. Z., Burns, B., Dubs, M. S., & Thompson, S. (2002). Assessing conformity to standards for treatment foster care. Journal of Emotional and Behavioral Disorders, 10(4), 213–222. Farmer, E. M. Z., Dorsey, S., & Mustillo, S. A. (2004). Intensive home and community interventions. Child and Adolescent Psychiatric Clinics of North America, 13, 857–884. Fisher, P. A., Ellis, B. H., & Chamberlain, P. (1999). Early intervention foster care: A model for preventing risk in young children who have been maltreated. Children Services: Social Policy, Research, and Practice, 2, 159–182. Fisher, P. A., Gunnar, M. R., Chamberlain, P., & Reid, J. B. (2000). Preventive intervention for maltreated preschool children: Impact on children’s behavior, neuroendocrine activity, and foster parent functioning. Journal of the American Academy of Child and Adolescent Psychiatry, 39(11), 1356–1364. Foster, E. M., & Connor, T. (2005). Public costs of better mental health services for children and adolescents. Psychiatric Services, 56, 50–55. Friedman, R. M. (1984). Seriously emotionally disturbed children: An underserved and ineffectively served population. Florida: Mental Health Institute. Friesen, B. J., Pullmann, M., Koroloff, N. M., & Rea, T. (2005). Multiple perspectives on family outcomes in children’s mental health. In: M. H. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: programs and evaluation best practices (pp. 101–123). Austin, TX: Pro-Ed Publishing.

The Evidence for Home and Community-Based Mental Health Services

175

Hahn, R. A., Bilukha, O., Lowy, J., Crosby, A., Fullilove, M. T., Liberman, A., Moscicki, E., Snyder, S., Tuma, F., Corso, P., & Schoﬁeld, A. (2005). The effectiveness of therapeutic foster care for the prevention of violence. American Journal of Preventive Medicine, 28(281), 72–90. Henggeler, S. W. (2003). Multisystemic therapy: An overview. Data, dissemination, and direction. Paper presented at the National Association of State Mental Health Program Directors Research Institute Conference. Henggeler, S. W., Borduin, C. M., Melton, G. B., Mann, B. J., Smith, L. A., Hall, J. A., Cone, L. T., & Fucci, B. R. (1991). Effects of multisystemic therapy on drug use and abuse in serious juvenile offenders: A progress report from two outcome studies. Family Dynamics of Addiction Quarterly, 1(3), 40–51. Henggeler, S. W., Clingempeel, W. G., Brondino, M. J., & Pickrel, S. G. (2002). Four-year follow-up of multisystemic therapy with substance-abusing and substance-dependent juvenile offenders. Journal of the American Academy of Child and Adolescent Psychiatry, 41(7), 868–874. Henggeler, S. W., Melton, G., Brondino, M. J., & Scherer, D. G. (1997). Multisystemic therapy with violent and chronic juvenile offenders and their families: The role of treatment ﬁdelity in successful dissemination. Journal of Consulting and Clinical Psychology, 65(5), 821–833. Henggeler, S. W., Melton, G., & Smith, L. A. (1992). Family preservation using multisystemic therapy: An effective alternative to incarcerating serious juvenile offenders. Journal of Consulting and Clinical Psychology, 60(6), 953–961. Henggeler, S. W., Melton, G. B., Smith, L. A., Schoenwald, S. K., & Hanley, J. H. (1993). Family preservation using multisystemic treatment: Long-term follow-up to a clinical trial with serious juvenile offenders. Journal of Child and Family Studies, 2(4), 283–293. Henggeler, S. W., Pickrel, S. G., & Brondino, M. J. (1999a). Multisystemic treatment of substance-abusing and -dependent delinquents: Outcomes, treatment, ﬁdelity, and transportability. Mental Health Services Research, 1(3), 171–184. Henggeler, S. W., Rodick, J. D., Borduin, C. M., Hanson, C. L., Watson, S. M., & Urey, J. R. (1986). Multisystemic treatment of juvenile offenders: Effects on adolescent behavior and family interaction. Developmental Psychology, 22(1), 132–141. Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C. A., Cunningham, P. B., & Pickrel, S. G. (2005). Predictors of treatment response for suicidal youth referred for emergency psychiatric hospitalization. Journal of Clinical Child and Adolescent Psychology, 34(3), 582–589. Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C. A., Sheidow, A., Ward, D. M., Randall, J., Pickrel, S. G., Cunningham, P. B., & Edwards, J. (2003). One-year follow-up of multisystemic therapy as an alternative to the hospitalization of youths in psychiatric crisis. Journal of the American Academy of Child and Adolescent Psychiatry, 42(5), 543–551. Henggeler, S. W., Rowland, M. D., Randall, J., Ward, D. M., Cunningham, P. B., Miller, S. L., Edwards, J., Zealberg, J. J., Hand, L. D., & Santos, A. B (1999b). Home-based multisystemic therapy as an alternative to the hospitalization of youths in psychiatric crisis: Clinical outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 38(11), 1331–1339. Henggeler, S. W., Schoenwald, S. K., Rowland, M. D., & Cunningham, P. B. (2002). Serious emotional disturbance in children and adolescents: Multisystemic therapy. New York: Guilford Press.

176

TRACY J. PINKARD AND LEONARD BICKMAN

Henggeler, S. W., Schoenwald, S. K., Swenson, C. P., & Borduin, C. M. (2006). Methodological critique and meta-analysis as Trojan horse. Children and Youth Services Review, 28, 447–457. Hoagwood, K., Burns, B. J., Kiser, L., Ringeisen, H., & Schoenwald, S. K. (2001). Evidence based practice in child and adolescent mental health services. Psychiatric Services, 52(9), 1179–1189. Huang, L., Stroul, B. A., Friedman, R. M., Mrazek, P. J., Friesen, B., Pires, S., & Mayberg, S. (2005). Transforming mental health care for children and their families. American Psychologist, 60(6), 615–627. Huey, S. J., Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C. A., Cunningham, P. B., Pickrel, S. G., & Edwards, J. (2004). Multisystemic therapy effects on attempted suicide by youths presenting psychiatric emergencies. Journal of the American Academy of Child and Adolescent Psychiatry, 43(2), 183–190. Institute of Medicine. (2001). Crossing the quality chasm: A new health system for the 21st century. Washington, DC: Institute of Medicine. Kazdin, A. E. (2004). Evidence-based treatments: Challenges and priorities for practice and research. Child and Adolescent Psychiatric Clinics of North America, 13(4), 923–940. Kazdin, A. E. (2006). Arbitrary Metrics: Implications for identifying evidence-based treatments. American Psychologist, 61(1), 42–49. Kielser, C. (1993). Mental health policy and the psychiatric inpatient care of children. Applied and Preventive Psychology, 2, 91–99. Knitzer, J. (1982). Unclaimed children. Washington, DC: Children’s Defense Fund. Lambert, E. W., Doucette, A., & Bickman, L. (2001a). Measuring mental health outcomes with pre-post designs. Journal of Behavioral Health Services and Research, 28(3), 273–286. Lambert, M. J., Hansen, N. B., & Finch, A. E. (2001b). Patient-focused research: Using patient outcome data to enhance treatment effects. Journal of Consulting and Clinical Psychology, 69(2), 159–172. Lambert, M. J., Ogles, B. M., & Masters, K. S. (1992). Choosing outcome assessment devices: An organizational and conceptual scheme. Journal of Counseling and Development, 70, 527–532. Lambert, M. J., Whipple, J. L., Smart, D. W., Vermeersch, D. A., Nielsen, M. E., & Hawkins, E. J. (2001c). The effects of providing therapists with feedback on patient progress during psychotherapy: Are outcomes enhanced. Psychotherapy Research, 11(1), 49–68. Leve, L. D., & Chamberlain, P. (2005a). Association with delinquent peers: Intervention effects for youth in the juvenile justice system. Journal of Abnormal Child Psychology, 33(3), 339–347. Leve, L. D., & Chamberlain, P. (2005b). Girls in the juvenile justice system: Risk factors and clinical implications. In: D. J. Pepler, K. C. Madsen, C. Webster & K. Levene (Eds), The development and treatment of girlhood aggression (pp. 191–215). Mahwah, NJ: Lawrence Erlbaum Associates. Littell, J. H. (2005). Lessons from a systematic review of effects of multisystemic therapy. Children and Youth Services Review, 27, 445–463. Littell, J. H. (2006). The case for multisystemic therapy: Evidence or orthodoxy? Children and Youth Services Review, 28, 458–472. Meadowcroft, P., Thomlison, B., & Chamberlain, P. (1994). Treatment foster care services: A research agenda for child welfare. Child Welfare, 73(5), 565. Munger, R. L. (1998). The ecology of troubled children. Cambridge, MA: Brookline Books.

The Evidence for Home and Community-Based Mental Health Services

177

Myaard, M. J., Crawford, C., Jackson, M., & Alessi, G. (2000). Applying behavior analysis within the wraparound process: A multiple baseline study. Journal of Emotional and Behavioral Disorders, 8(4), 216–229. Posavac, E. J., & Carey, R. G. (2003). Program evaluation: Methods and case studies (6th ed.). Upper Saddle River, NJ: Prentice Hall. President’s Commission on Mental Health. (1978). Report on the sub-task panel on infants, children and adolescents. Washington, DC: U.S. Government Printing Ofﬁce. Pullmann, M. D., Kerbs, J., Koroloff, N., Veach-White, E., Gaylor, R., & Sieler, D. (2006). Juvenile offenders with mental health needs: Reducing recidivism using wraparound. Crime and Delinquency, 52, 375–397. Riemer, M., Rosof-Williams, J., & Bickman, L. (2005). Theories related to changing clinician practice. Child and Adolescent Psychiatric Clinics of North America, 14, 241–254. Ringel, J., & Sturm, R. (2001). National estimates of mental health utilization and expenditures for children in 1998. Journal of Behavioral Health Services and Research, 28(3), 319–333. Rosie D. et al. v. Mitt Romney et al. (2006). Rowland, M. D., Halliday-Boykins, C. A., Henggeler, S. W., Cunningham, P. B., Lee, T. G., Kruesi, M. J., & Shapiro, S. B. (2005a). A randomized trial of multisystemic therapy with Hawaii’s Felix Class youths. Journal of Emotional and Behavioral Disorders, 13(1), 13–23. Rowland, M. D., Halliday-Boykins, C. A., & Schoenwald, S. K. (2005b). Multisystemic therapy with youth exhibiting signiﬁcant psychiatric impairment. In: M. H. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: Programs and evaluation best practices (pp. 401–419). Austin, TX: Pro-Ed Publishing. Sapyta, J., Riemer, M., & Bickman, L. (2005). Feedback to clinicians: Theory, research, and practice. Journal of Clinical Psychology, 61(2), 145–153. Schoenwald, S. K., Sheidow, A., & Letourneau, E. (2004). Toward effective quality assurance in evidence-based practice: Links between expert consultation, therapist ﬁdelity, and child outcomes. Journal of Clinical Child and Adolescent Psychology, 33(1), 94–104. Schoenwald, S. K., Ward, D. M., Henggeler, S. W., Pickrel, S. G., & Patel, H. (1996). Multisystemic therapy treatment of substance abusing or dependent adolescent offenders: Costs of reducing incarceration, inpatient, and residential placement. Journal of Child and Family Studies, 5(4), 431–444. Schoenwald, S. K., Ward, D. M., Henggeler, S. W., & Rowland, M. D. (2000). Multisystemic therapy versus hospitalization for crisis stabilization of youth: placement outcomes 4 months postreferral. Mental Health Services Research, 2(1), 3–12. Sexton, T., & Alexander, J. F. (2003). Functional family therapy: A mature clinical model for working with at-risk adolescents and their families. In: T. Sexton, G. Weeks & M. Robbins (Eds), Handbook of family therapy: The science and practice of working with families and couples (pp. 323–348). New York: Brunner-Routledge. Sheidow, A. J., Bradford, W. D., Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C. A., Schoenwald, S. K., & Ward, D. M. (2004). Treatment costs for youths receiving multisystemic therapy or hospitalization after a psychiatric crisis. Psychiatric Services, 55(5), 548–554. Shirk, S. R. (2004). Dissemination of youth ESTs: Ready for prime time? Clinical Psychology: Science and Practice, 11(3), 308–312. Silverman, W. K., Kurtines, W., & Hoagwood, K. (2004). Research progress on effectiveness, transportability, and dissemination of empirically supported treatments: Integrating theory and research. Clinical Psychology: Science and Practice, 11(3), 295–299.

178

TRACY J. PINKARD AND LEONARD BICKMAN

Smith, D. K. (2004). Risk, reinforcement, retention in treatment, and reoffending for boys and girls in multidimensional treatment foster care. Journal of Emotional and Behavioral Disorders, 12(1), 38–48. Smith, D. K., Stormshak, E., Chamberlain, P., & Whaley, R. B. (2001). Placement disruption in treatment foster care. Journal of Emotional and Behavioral Disorders, 9(3), 200–205. Society for Prevention Research. (2004). Standards of evidence: Criteria for efficacy, effectiveness and dissemination. Falls Church, VA: Society for Prevention Research. Stortz, M. (2000). County Mental Health responsibility to provide community services. Stroul, B. A., & Friedman, R. M. (1986). A system of care for children and youth with severe emotional disturbances. Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Stroul, B. A., & Friedman, R. M. (1994). A system of care for children and youth with severe emotional disturbances (Rev. ed). Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Substance Abuse and Mental Health Services Administration (SAMHSA). (2005). (FY) 2007 Justiﬁcation of Estimates for Appropriations Committees. Sutphen, R. D., Thyer, B. A., & Kurtz, P. D. (1995). Multisystemic treatment of high-risk juvenile offenders. International Journal of Offender Therapy and Comparative Criminology, 39(4), 329–334. Tierney, J. P., Grossman, J. B., & Resche, N. L. (1995). Making a difference: An impact study of Big Brothers/Big Sisters. Philadelphia, PA: Public/Private Ventures. U.S. Congress Ofﬁce of Technology Assessment. (1986). Children’s mental health: Problems and services – A background paper. Washington, DC: U.S. Government Printing Ofﬁce. U.S. Department of Health and Human Services. (1999). Mental health: A report of the surgeon general. Rockville, MD: Department of Health and Human Services, Substance Abuse and Mental Health Administration, Center for Mental Health Services, National Institutes of Health, National Institutes of Mental Health. U.S. Public Health Service. (2000). Report of the surgeon general’s conference on children’s mental health: A national action agenda. Washington, DC: Department of Health and Human Services. Warren, R., & Cohen, S. (1985). Respite care. Rehabilitation Literature, 46, 66–71. Weisz, J. R., Sandler, I. N., Durlak, J. A., & Anton, B. S. (2005). Promoting and protecting youth mental health through evidence-based prevention and treatment. American Psychologist, 60(6), 628–648.

CHALLENGES FOR A SYSTEM OF CARE Kathleen Biebel and Jeffrey L. Geller ‘‘System of care’’ has become the buzzword for contemporary child and adolescent services. Despite the fact that ﬁndings of studies of child and family outcomes associated with system of care have been disappointing (Evans & Armstrong, 2005), systems of care are at the forefront of service delivery for children and adolescents with serious emotional disturbances (SED). Systems of care, however, are not without controversy. This chapter will explore various macro and micro level challenges currently facing the system of care movement. At the macro and systems level, challenges include meeting the needs of the high-risk population of children and adolescents traditionally served by systems of care, and securing cooperation and collaboration from multiple child serving systems. At the micro and practice level, challenges include ﬁnancing systems of care, shortages of mental health services and staff, and incorporating inpatient psychiatric care into a community-based treatment model. These challenges are highlighted by ﬁndings from empirical investigations of systems of care and other interventions for children and adolescents with SED. Finally, future implications for systems of care will be discussed.

SYSTEM OF CARE A system of care is a function-speciﬁc, rather than agency-speciﬁc approach deﬁned as a ‘‘comprehensive spectrum of mental health and other necessary Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 179–199 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14009-X

179

180

KATHLEEN BIEBEL AND JEFFREY L. GELLER

services which are organized into a coordinated network to meet the multiple and changing needs of children and adolescents with severe emotional disturbances and their families’’ (Stroul & Friedman, 1986). A system of care provides a mental health delivery system for children with SED with a wide array of accessible, community-based services that focus on children’s individual needs, include the family in treatment planning, and provide culturally competent services. System of care services are provided by multiple child serving agencies and are collaborative and coordinated (Stroul & Friedman, 1986). The philosophy and values underlying systems of care include: focus on the individual child and family; provision of services within the most normalized environment that meets the youth’s needs; creation of partnerships with families; use of a strength-based, ecological orientation; creation of an environment of cultural competence; and provision of unconditional care (Lourie, Katz-Levy, & Stroul, 1996; Stroul & Friedman, 1986). A system of care framework typically consists of mental health interventions, social and educational services, healthcare, substance abuse interventions, recreation, and operational services (Stroul & Friedman, 1986). Care and treatment in the mental health component of a system of care have both nonresidential and residential services. Nonresidential services include: assessment, psychiatric treatment, outpatient interventions (individual, family, and group), home-based services, day treatment, crisis intervention, after-school and evening programs, therapeutic respite, behavioral aides, case management, and parent education and support services. Residential services include: therapeutic foster care, therapeutic group homes, crisis residences, residential treatment, and inpatient hospitalization (Stroul, McCormack, & Zaro, 1996). All components of the system of care are interrelated and the effectiveness of any one component is related to the availability and the effectiveness of all other components (Stroul & Friedman, 1986). There must be an appropriate balance between the components, particularly between the more restrictive and less restrictive services (Stroul & Friedman, 1986). At this time, however, there are no evidence-based guidelines as to the appropriate/medically necessary capacity of each component of a system of care (Stroul & Friedman, 1986). Outcome measures are necessary to evaluate systems of care. Outcome measurements applied to systems of care include the effects on symptoms and diagnoses (Hoagwood, Jensen, Petti, & Burns, 1996), the use of outof-home and out-of-community placements, the utilization of restrictive service options, the functioning of youngsters, the educational status of youngsters, the law enforcement status of youngsters, family involvement,

Challenges for a System of Care

181

satisfaction with services, access to services, costs (Cook & Kilmer, 2004; Jensen, Hoagwood, & Petti, 1996; Stroul et al., 1996) and standardized test results (Greenbaum et al., 1996). Systems of care are most often envisioned as a continuum – a linear progression of services based on intrusiveness or restrictiveness. Another way to view systems of care is as a spectrum of services – an array of services from which the most appropriate form of treatment/care is selected. A study of children’s services in Philadelphia demonstrated that the spectrum model, rather than the continuum model, most closely deﬁnes what actually takes place (Dore, Wilkinson, & Sonis, 1992). This ﬁnding mirrors results found earlier for adults (Geller & Fisher, 1993). More recently, investigators have noted that little is known about how youths move into and through various components of systems of care (Farmer, Burns, Phillips, Angold, & Costello, 2003). There is empirical evidence, however, that experienced clinicians can and will appropriately direct child patients to alternative treatment settings based on ‘‘an algorithm involving the severity of psychopathology and the presence of several vulnerable factors’’ when provided the opportunity to do so using referral and assessment data (McDermott, McKelvey, Roberts, & Davies, 2002). In response to the development of systems of care, the American Psychiatric Association (APA) promulgated a report on quality indicators for psychiatric care for children and adolescents (American Psychiatric Association [APA] Task Force, 2002). The report includes the following observations and recommendations: (1) Children and adolescents presenting with symptoms of behavioral, emotional, or learning difﬁculties need an assessment that is sufﬁciently comprehensive, linguistically and culturally sensitive, and of adequate duration, to establish or rule out the presence of a psychiatric disorder, to make an appropriate DSM-IV-TR diagnosis, and to determine the level of impairment. (2) Interventions of varying levels of intensity and restrictiveness should be available as appropriate for those disorders for which they have been shown to be efﬁcacious, effective, or clinically indicated. (3) Children and adolescents with mental illness should receive care and treatment in the least restrictive/most normative settings that are appropriate to their clinical needs. There are situations when out-of-home care is necessary. This recommendation is not meant to unnecessarily limit clinically necessary and appropriate out-of-home care, but rather that such loci should be used advisedly.

182

KATHLEEN BIEBEL AND JEFFREY L. GELLER

The report also indicates that the evidence basis for inpatient treatment and for community-based alternatives for children and adolescents is limited.

CHALLENGE I: SYSTEMS OF CARE SERVE CHILDREN AND ADOLESCENTS MOST IN NEED Children and adolescents enrolled in system of care services are generally the least diagnostically simple cases – they have multiple biological, familial, and socioeconomic vulnerabilities. Analyses of data from the SAMHSA National Evaluation of the Comprehensive Community Mental Health Services for Children and Their Families Program found that over 60% of children with serious emotional disturbance entering system of care services had a history of one or more risk factors (Center for Mental Health Services [CMHS], 1999). These data indicate that 30% had histories of running away, 26% had a previous psychiatric hospitalization, 24% had been physically abused, 20% had a history of substance abuse, and 18% had been sexually abused (CMHS, 1999). While children and adolescents with SED are likely to be exposed to various risk factors and face multiple challenges throughout childhood, the current me´lange of systems, agencies, and services may actually be exacerbating the serious and emotional disturbances they are or are not set up to address. Signiﬁcant attention in the research literature has been paid to the increased likelihood of incarceration, co-occurring mental health and substance abuse disorders, suicide, and school failure. And while the majority of children and adolescents improve during inpatient treatment, there are some, particularly those with strong antisocial behavior, who appear to beneﬁt less (Cornsweet, 1990). Youth in the juvenile justice system experience signiﬁcantly higher rates of mental health disorders than youth in the general population (Cocozza & Skowyra, 2000), and more than half have co-occurring substance abuse disorders (Goldstrom, Jaiquan, Henderson, Male, & Manderscheid, 2001). A 2003 report by the General Accounting Ofﬁce identiﬁed inappropriate incarceration of youth with mental disorders as one of the most severe consequences of insufﬁcient mental health care (United State House of Representatives Committee, 2004). Incarcerated youths generally get either no or inadequate services as ‘‘traditional services in the juvenile justice system tend to either ignore mental health needs or inappropriately treat youths with multifaceted problems’’ (Rosenblatt, Rosenblatt, & Biggs, 2000).

Challenges for a System of Care

183

The rates of co-occurring mental health and substance abuse disorders among children and adolescents are high. The National Comorbidity Study found that 41–65% of individuals with a lifetime substance abuse disorder also had a lifetime history of at least one psychiatric disorder, and that 51% of those with at least one lifetime psychiatric disorder also had at least one substance abuse disorder; these rates were highest among 15–24 year olds (Kessler et al., 1994). Research indicates that of youth 11 years and older with SED, over half had tried alcohol (54%) and cigarettes (67%) at least once while 43% had tried marijuana at least one time (CMHS, 1999). Making the issue of dual diagnoses even more complex is that children and adolescents experiencing mental health problems may employ drugs and alcohol as a means to self-medicate (Danielson, Overholser, & Butt, 2003; Kelly et al., 2003). Annual suicide attempts in 15–24 year olds have surpassed one million in the United States (Greenﬁeld, Larson, Hechtman, Rousseau, & Platt, 2002) and for this age group suicide is the third leading cause of death. Research indicates that boys are four times more likely to commit suicide than girls, but girls are twice as likely as boys to attempt suicide (United States Department of Health and Human Services, 1999). Adolescents who make multiple suicide attempts are three times more likely to die by suicide than are adolescents who make one attempt (Brinkman-Sull, Overholser, & Silverman, 2000). From the early 1960s to the early 1990s, the United States has seen an astounding 194% increase in the suicide rates for persons younger than 15 years of age (DeLeo & Evans, 2004). Children and adolescent inpatient admissions have seen an increase not only in suicidal behaviors, but also in assaultive behaviors (Cornsweet, 1990). Research indicates that children with untreated mental health disorders have great difﬁculties both learning adequately in school and developing the interpersonal skills necessary to lead productive adult lives. Almost half of students with mental disorders drop out of grades 9–12, and of those dropouts 73% are arrested within 5 years of leaving school (APA, 2002).

CHALLENGE II: SECURING COOPERATION AND COLLABORATION OF MULTIPLE CHILD SERVING SYSTEMS The majority of high-risk children and adolescents with SED receiving mental health care interact with multiple service systems because alltoo-often the needs of children and adolescents span across different service

184

KATHLEEN BIEBEL AND JEFFREY L. GELLER

sectors. For example, if there are issues of caregiver abuse and neglect the child with SED may be involved with the child welfare system, or if a child with SED has been involved in criminal activity he or she may have contact with the juvenile justice system. Given that the mental health system is often ill-equipped to provide for its own population, it is no surprise that these sister systems are equally challenged to provide for the mental health needs of their constituents. Often other service systems are called upon to ‘‘pick up the slack’’ of the overwhelmed children’s mental health system. But these service systems also have limited resources and often lack the specialized training necessary to attend to the mental health concerns of children and adolescents with emotional and behavioral problems assigned to their care (Bazelon Center for Mental Health Law, 2004). The juvenile justice system is one example of a child serving system often involved in a system of care. Trends in the juvenile justice system mirror those of the adult criminal system: an increased reliance on the justice system to care for individuals with mental illness and a movement away from rehabilitation and toward punitive measures. The mental health needs of youth in the juvenile justice system have received more attention at the federal level in the past two years than in the past three decades combined (Cocozza & Skowyra, 2000). The U.S. Department of Health and Human Services Center for Mental Health Services recently initiated the ﬁrst national survey of juvenile justice facilities in an effort to identify available mental health services. At this time, vast numbers of children and adolescents are being inappropriately incarcerated due to a lack of mental health resources. ‘‘Juvenile detention centers are designed to care for children who have been charged with crimes and those who are awaiting court hearings or placement. Juvenile detention facilities lack the resources and staff to confront this problem; yet, corrections is being forced to shoulder the burden of the nation’s failure to properly diagnose and care for children with mental or emotional disorders’’ (Bazelon Center, 2004).

CHALLENGE III: FINANCING SYSTEMS OF CARE Many of the services essential to systems of care do not have readily available funding streams. States are reporting difﬁculty securing reliable and stable ﬁnancing for services such as respite, mentoring, and parent-to-parent support, which make up the foundation of wraparound services within a system of care (Biebel & Katz-Leavy, 2005). Many services most helpful to children with SED could be reimbursed by Medicaid but rarely are.

Challenges for a System of Care

185

Medicaid is federally mandated to pay for all medically necessary services authorized by federal law for children, which include inpatient hospital treatment for children under 21 years old, out-of-home treatment centers, clinic services, outpatient hospital services, prescription drugs, and rehabilitation services. It is left to the discretion of states to deﬁne these services. This can create problems for providers in terms of knowing how to bill Medicaid or even what services are reimbursable. This lack of clarity can signiﬁcantly affect the availability of any particular service for a child. Many of the more intensive community-based services, which families report as most helpful to them, are not traditionally covered by states in their Medicaid Plans (Bazelon Center for Mental Health Law, 1999). These may include, but are not limited to, crisis services, intensive in-home services, substance abuse counseling, social and daily living skills training, case management, behavioral aide services, and other intensive communitybased care. The trend in the health insurance arena toward managed care may be even more problematic for children and adolescents with SED difﬁculties than for adults. Managed care organizations often lack the specialized resources, e.g., day treatment and long-term services, required for high-need children and adolescents. Other concerns about managed behavioral health care include limited outpatient visits, inadequate emphasis on communitybased services, eschewal of psychosocial rehabilitation interventions, insufﬁcient evidence regarding managed care-sanctioned therapeutic interventions, and the actual cost-effectiveness of managed care techniques themselves (Shera, 1996). More vulnerable populations may not be well served by a behavioral managed care system, and the costs of serving the severely ill may discourage HMOs from providing high-quality mental health services at all (Callahan & Merrick, 1997; Scallet, 1996). Managed care strategies may be inappropriate for delivering mental health services to the most seriously ill (Geller, 1998). Restrictions inherent in behavioral managed care may not allow for the ﬂexibility of services required by many children and adolescents.

CHALLENGE IV: SHORTAGES OF MENTAL HEALTH SERVICES AND STAFF Services and treatment options for children and adolescents with mental health needs are limited at all points along the continuum of care. A 2000 study conducted by the Children’s League of Massachusetts documents a

186

KATHLEEN BIEBEL AND JEFFREY L. GELLER

critical shortage of services for children with mental health concerns (Massachusetts Citizens for Children, 2001). At the same time, children become stuck in psychiatric wards and mental health units as they await referrals to other therapeutic settings. Many children in need of specialized inpatient level of care are refused admittance to child psychiatric wards due to a shortage of beds. Oftentimes children remain for extended periods of time in emergency rooms waiting for an appropriate bed to become available, a practice which can compound children’s fragile state. ‘‘Children are presenting more serious levels of pathology, their numbers are increasing, and their opportunities for treatment and recovery are diminishing due to the shortage of beds and lack of qualiﬁed treatment providers’’ (Massachusetts Citizens for Children, 2001). Reports and studies from across the USA have detailed the shortcomings of states’ children’s mental health systems. A 2001 study by the Washington State Emergency Medical Services for Children group noted a signiﬁcant lack of mental health services for children and adolescents. At the systems level, major ﬁndings included: a lack of inpatient psychiatric beds for children and adolescents; inadequate mental health services to support a community-based model of mental health care; too few outpatient mental health services; signiﬁcant time delay, due to lack of resources, for child and adolescent evaluations; lack of private mental health providers; and limited resources for children and adolescents with dual diagnoses of mental health and alcohol/substance abuse disorders. Hospital speciﬁc concerns included: a shortage of mental health staff, speciﬁcally social workers and their services; hospital emergency departments with limited knowledge of mental health referrals for children and adolescents; a lack of emergency department mental and behavioral health screening tools; and a lack of clarity about emergency department’s role in identiﬁcation of nonacute mental and behavioral health concerns. The Lewin Group concluded after studying children’s services that ‘‘many children are not receiving the care they need from mental health specialists’’ (Scallet, Bush, Rockwood, & Fields, 2000). Researchers at RAND concluded that ‘‘the majority of children who are likely to beneﬁt from mental health services do not receive any care’’ (Ringel & Sturm, 2001). Shortages of general and specialized hospital staff have been reported for the last decade. A report commissioned by the California Institute for Mental Health describes signiﬁcant stafﬁng shortages, particularly among nurses and child psychiatrists (California Institute for Mental Health, 2001). The report highlights inpatient units as well as residential facilities having difﬁculties securing qualiﬁed staff especially for children’s programs. The

Challenges for a System of Care

187

2001 Surgeon General’s Report, the American Academy of Child and Adolescent Psychiatry (AACAP) and the National Association of Psychiatric Health Systems have each noted a severe shortage of psychiatrists and a dearth of other child-trained mental health professionals (Challenges facing behavioral health care, 2003; Kanapaux, 2004). Child-inpatient units are the most difﬁcult to staff, as children’s units require a higher staff-to-patient ratio than adult units. Suggested factors contributing to the shortages of child-trained professionals include rising housing costs, static salaries, the implementation of managed care, and opportunities in other industries. Given the high demand for child-trained professionals, they can easily shift to more highly paid jobs in other arenas and can ‘‘name their price,’’ a prospect that the limited resources of the children’s mental health system must confront.

CHALLENGE V: INPATIENT CARE AS PART OF A SYSTEM OF CARE Psychiatric hospitals, often referred to as ‘‘mental hospitals,’’ have been seen as ‘‘snake pits’’ for decades, particularly the public hospitals (Grob, 1994). At their peak census, in mid-20th century, when public psychiatric hospitals had 50% of all hospital beds (Gorman, 1956) these hospitals were labeled ‘‘the shame of the states’’ (Deutsch, 1948). To a signiﬁcant degree the bad taste left by mental hospitals has never gone away. And there is little doubt that these perceptions have tainted opinions about child inpatient psychiatry. For example, in 2004, a professional journal published an article titled, ‘‘Child psychiatric hospitalization: The last resort’’ (Scharer & Jones, 2004). In the contemporary era of the privatization of psychiatric inpatient treatment, the private hospitals became as villainized as the public hospitals, if not more so, through egregious practices aimed at ﬁlling child and adolescent beds. Hospitalizing adolescents became a market-driven phenomenon, all too often ignoring medical needs. Deviance was medicalized and private hospitals were accused of admitting any child or adolescent for whom insurance reimbursement was available (Sharky, 1994). As one commentator pointed out in the early 1990s, ‘‘we hospitalize our troublesome youth, instead of our troubled youth’’ (Kiesler, 1993). Fundamental to an understanding of systems of care is that no system of care has functioned without the availability of a full range of services, with one component of that range of services being inpatient hospital level of

188

KATHLEEN BIEBEL AND JEFFREY L. GELLER

care. No knowledgeable person has written that systems of care do not need inpatient hospital capacity. To the contrary, the innovators of systems of care and their followers have underscored the need for the availability of inpatient treatment in the system. Within a system of care, inpatient hospitalization provides four critical services: short-term treatment and crisis stabilization; comprehensive evaluation; long-term treatment; and specialized services (Stroul & Friedman, 1986). While some states have technically closed all their public child and adolescent beds, they often fund, and sometimes even directly run, what is tantamount to state child and adolescent psychiatric inpatient beds. The authors have directly observed that Massachusetts, for example, directs funds from the state mental health authority to the state university so that the university can operate intensive residential treatment beds. These beds are located on a state hospital campus in a locked building formerly used for adult patients. A visitor to this unit, or an adolescent on the unit, would have few experiences qualitatively different from a visit or admission to the state psychiatric hospital for children and adolescents of a decade ago. States need to use caution when making comparisons across states; beware the state that advocates closing all its public beds using as support the belief another state has no public child and adolescent beds. Professional associations have expressed concern about the inadequacy of and reluctance to use child and adolescent inpatient beds. It has been the position of the APA since 1989 that there is a shortage of inpatient treatment facilities for children and adolescents, and that ‘‘large numbers of children and adolescents with diagnosable disorders are not receiving the inpatienty treatment that could help them’’ (APA, 1989). In 1989, the AACAP issued a policy statement, ‘‘Inpatient Hospital Treatment of Children and Adolescents’’ indicating the Academy supports the use of inpatient psychiatric treatment in a hospital setting when the psychiatric needs of a child or adolescent, as assessed by a properly qualiﬁed psychiatrist, warrant such treatment and when the treatment provided is of high quality (American Academy of Child and Adolescent Psychiatry [AACAP], 1989). One recent recommendation indicated, ‘‘Both long- and short-term hospital stays should be available, and decisions made between them based on which is most appropriate for a particular client’s situation. Many children and adolescents need only crisis interventiony other children and teenagers, however, need the longer term care of a nurturing, structured, therapeutic and supportive environment’’ (Molidor, 1997). Using a database that integrated information for mental health treatment, substance abuse services and Medicaid, researchers concluded in a three-state study that their results

Challenges for a System of Care

189

substantiate the importance of inpatient psychiatric care for the population of children with SED in the public sector (Teich, Buck, Graver, Schroeder, & Zheng, 2003). While individuals from all disciplines in the mental health ﬁeld are providing better data to support when and when not to hospitalize children and adolescents, persons in the legal advocacy community continue to proselytize in disjointed ways. One Chicago attorney opined, ‘‘Children should only be institutionalized in mental health hospitals when no less restrictive alternative exists to cure their psychological problems’’ (italics added) (Tsesis, 1998). There is evidence that inpatient child and adolescent psychiatry is improving. Psychiatric units/hospitals have become more attuned to the needs and sensibilities of children with severe mental health disorders (2003 APA Gold Award). Efforts to use less intrusive, less restrictive interventions in inpatient units/hospitals for children and adolescents is in evidence, particularly through efforts at seclusion and restraint reform (Donovan, Siegel, Zera, Plant, & Martin, 2003). While many children and adolescents need only crisis intervention, other youth may require longer term care in a structured, therapeutic, and supportive environment. The AACAP supports the use of inpatient psychiatric treatment in a hospital setting when the psychiatric needs of a child or adolescent, as assessed by a properly qualiﬁed psychiatrist, warrant such treatment and when the treatment provided is of high quality. Recommendations for the use of psychiatric inpatient treatment state that both longand short-term hospital stays should be available. In 1996, the APA and the AACAP issued criteria for Short-Term Treatment of Acute Psychiatric Illness. Components of this document relevant to inpatient treatment indicate clinical criteria for acute inpatient hospital admission, for continued treatment, and for continued care. While insurance continues to play a role in which children and adolescents are hospitalized (Pottick, Hansell, Gutterman, & White, 1995), recent studies show much progress in deﬁning an effective role for inpatient treatment of children and adolescents. Recent research has delineated who gets hospitalized and under what circumstances (Bickman, 1996; Heﬂinger, Simpkins, & Foster, 2002; Lyons, Kisiel, Dulcan, Cohen, & Chesler, 1997; Strauss, Chassin, & Lock, 1995); factors inﬂuencing rehospitalization (Arnold et al., 2003; Bladder, 2004; Romansky, Lyons, Lehner, & West, 2003); variables inﬂuencing length of stay (Hoger et al., 2002); and outcomes (Mayes, Calhoun, Krecko, Vesell, & Hu, 2001; Sourander & Piha, 1998). Perhaps the most promising ﬁnding is that there is a rational process directing hospital admissions wherein the majority of decisions to

190

KATHLEEN BIEBEL AND JEFFREY L. GELLER

hospitalize children were supported by each child’s level of risk (Lyons et al., 1997). Satisfaction expressed by patients and parents is being examined for inpatient treatment of children and adolescents. One study, done in New York State at an 80-bed free-standing, state children’s psychiatric hospital found a high level of satisfaction with patient care in a study of 157 inpatients and 111 parents or guardians. Results indicated a rate of 95% satisfaction expressed by the child patients and 87% satisfaction indicated by parents (Kaplan, Busner, Chibnail, & Kang, 2001). Systems of care are multifaceted programs with national, generalizable features and unique, local modiﬁcations. It is clear that to date, even the best systems of care require inpatient psychiatric treatment settings for some youth whose needs are beyond those community-based resources can address. The literature on system of care seems to indicate that forcing a system of care to function without one key component of the treatment continuum – inpatient psychiatric hospital beds – will tax the system of care beyond its capacity, failing not only those who need more intense services than the community can deliver but also those whose needs go unmet as the system of care crumbles due to the absence of a fundamental component. Empirical investigations and program descriptions have repeatedly showed the necessity for inpatient level of care in systems of care.

EMPIRICAL INVESTIGATIONS OF SYSTEM OF CARE The ﬁrst and best-known empirical evaluation of a system of care is the Fort Bragg (Texas) study (Bickman, 1996; Bickman, Foster, & Lambert, 1996; Bickman, Lambert, Andrade, & Penaloza, 2000; Bickman, Summerfelt, & Foster, 1996). In this study military dependents in the Fort Bragg area were provided with a broad range of services not the least of which were a single point of entry, comprehensive assessments, and no co-payment or beneﬁt limit. The impact of this system of care was determined by comparing the children of Fort Bragg with children receiving standard services through CHAMPUS insurance at two other southeastern military bases. Major outcomes included: (1) systems reforms can be successfully implemented with sufﬁcient resources (Bickman, 1996); (2) these reforms can increase access and signiﬁcantly improve satisfaction (Bickman, 1996); (3) systems of care had little effect on important clinical and functional outcomes (Bickman et al., 2000); (4) clinicians used alternatives to hospitalization where these alternatives existed and could be accessed (Bickman et al., 1996);

Challenges for a System of Care

191

(5) hospitalization was still utilized for the most severe cases, even in a fully developed system of care (Bickman et al., 1996); and (6) ‘‘the notion that costs can be controlled by clinicians and their managers by placing children in what they believe to be the most ‘appropriate’ services for the ‘appropriate’ length of time was not supported’’ (Bickman, 1996). The Fort Bragg experiment was not without its critics (Bickman et al., 1998; Mordock, 1997, 1998). Those who conducted the Fort Bragg study refuted the criticism (Bickman et al., 1998) and extended their work by studying a system of care created in the public sector, working this time in Stark County, Ohio (Bickman, Andrade, & Lambert, 2002; Bickman, Noser, & Summerfelt, 1999; Bickman, Summerfelt, & Noser, 1997). Children and families were randomly assigned to immediate eligibility in a community-based system of care or to seek services on their own. In the ﬁrst observations of the outcomes, the researchers reported that those in the system of care received more case management services and more home visits than those in the comparison group, but there was no difference in clinical or functional outcomes (Bickman et al., 1997). Extending their ﬁndings to 18-month follow-up did not change the results (Bickman et al., 1999). Remarkably, there did not appear to be any relationship between the amount of services received and the outcomes (Bickman et al., 2002). To account for their ﬁndings, the authors posited that ‘‘the logic chain between system reform and clinical outcomes is too long’’; ‘‘the ability to identify and assign youths to appropriate services is not sufﬁciently well developed’’; the documentation of the efﬁcacy of psychotherapy in community treatment is inadequate; ‘‘there is not a consistent body of evidence that suggests that many of the innovative community-based treatments, such as home-based treatment or day treatment, are effective’’; and ‘‘other assumptions that underline system reform may also be inaccurate’’ (Bickman et al., 1999).

EMPIRICAL INVESTIGATIONS OF OTHER INTERVENTIONS FOR CHILDREN AND ADOLESCENTS WITH SED Multisystemic therapy (MST) was originally developed as a treatment intervention for youth in the juvenile justice system. A group at the Family Services Research Center at the University of South Carolina has spent much of the last decade studying MST (Henggeler, 1999; Henggeler et al., 1999, 2003; Huey et al., 2004; Schoenwald, Ward, Henggeler, & Rowland,

192

KATHLEEN BIEBEL AND JEFFREY L. GELLER

2000; Sheidow et al., 2004). The principles of MST, which are consistent with the system of care philosophy, are: (1) the primary purpose of assessment is to understand the ﬁt between the identiﬁed problems and their broader systemic context; (2) therapeutic contacts should emphasize the positive and should use systemic strengths as levers for change; (3) interventions should be designed to promote responsible behavior and decrease irresponsible behavior among family members; (4) interventions should be present-focused and action-oriented, targeting speciﬁc and well-deﬁned problems; (5) interventions should target sequences of behavior within or between multiple systems that maintain identiﬁed problems; (6) interventions should be developmentally appropriate and ﬁt the developmental needs of the youths; (7) interventions should be designed to require daily or weekly effort by family members; (8) intervention effectiveness is evaluated continuously from multiple perspectives, with providers assuming accountability for overcoming barriers to successful outcomes; and (9) interventions should be designed to promote treatment generalization and long-term maintenance of therapeutic change by empowering care-givers to address family members’ needs across multiple systemic contexts. (Henggeler, 1999)

In MST, there are low caseloads (ﬁve families per clinician), services are available 24 hours/day and 7 days/week for 4–6 months, and are strengthbased, individualized, and provided in families’ homes and other community locations. As with systems of care, families are seen as, and used as, valuable resources (Henggeler, 1999). Outcomes of particular relevance include: MST could decrease use of hospital days at times of crisis (Henggeler et al., 1999) and could sustain this level of dependence on inpatient treatment throughout the intervention period (Schoenwald et al., 2000) and beyond (Henggeler et al., 2003); MST was more cost effective than usual inpatient treatment and community aftercare (Sheidow et al., 2004); and MST could impact on decreasing suicide attempts signiﬁcantly more effectively than hospitalization (Huey et al., 2004). Noted in all the MST studies, however, is that some youths in MST could not be treated using only community resources and required inpatient treatment (Henggeler, 1999; Henggeler et al., 1999, 2003; Huey et al., 2004; Schoenwald et al., 2000; Sheidow et al., 2004). As stated in the most recent research effort, ‘‘it is important to note that 44% of MST youth were also admitted for psychiatric hospitalization during the course of treatment due to emergencies that could not be handled in community settings’’ (Huey et al., 2004). There is no other system of care or system of care-like interventions that have been as extensively studied as those above. The literature does report on others, however. A randomized trial of home-based family intervention versus routine care for children aged 10–16 years old in the United Kingdom showed the home-based intervention to be no more effective even with the ﬁnding that compliance with home-based treatment was better than with

Challenges for a System of Care

193

routine care (Harrington et al., 1998). Crisis services were studied in the Bronx, New York by randomly assigning children 5–17 years old to homebased crisis intervention, enhanced home-based crisis intervention, or crisis case management. Gains were made by all children, but were not differentiated by the type of services delivered. Of note is the authors’ reporting that ‘‘5% to 10% of the children enrolled in the in-home services did require hospitalization during the intervention period. These in-home programs did not eliminate the necessity of hospitalization for some children who posed a danger to themselves’’ (Evans, 2003; Evans, Boothroyd, & Armstrong, 1997). A national evaluation of the Comprehensive Community Mental Health for Children and Their Families Programs, in a study of 2,580 children (mean age was 12.1 years old) for whom there was at least 24 months of service, 45–50% of children showed clinical and functional improvement (Manteuffel, Stephens, & Santiago, 2002). These data reﬂect a cornucopia of systems of care and no comparison groups, and hence are hard to interpret.

IMPLICATIONS Throughout most of the history of American psychiatry, change has occurred through therapeutic ripples and ideological waves. Examining the progress of putative treatment for adults during the 20th century illustrates the unabashed endorsement of organic treatments, e.g., oophorectomies; insulin comas, and prefrontal lobotomies (Valenstein, 1986); psychoanalysis (Polnick, 1998); dehospitalization (mislabeled deinstitutionalization) (Geller, 2000); and psychopharmacologic interventions for all (Healy, 2004). For children and adolescents we have, with pendular regularity, waxed and wanted in our enthusiasm for integration versus substantially separate as the most effective way to attend to those with SED (Silver et al., 1992); we have built up and then taken down child psychiatric hospitals (Geller & Biebel, Part I, in press); and we have pushed the use of psychopharmacologic intervention, untested in children and adolescents, for all manner of psychiatric disorders, behavioral aberrations, and social acting out (Geller & Biebel, Part II, in press). This history looks more like a chronology of movements than it does like a history of science. Too much of American psychiatry is the tale of untested assumptions, one-size-ﬁts-all thinking, and jumping on bandwagons. And systems of care are at risk of being another blip in this sad history. As documented, efforts to establish an evidence basis for systems of care have generally had negative outcomes. Systems of care, in the most methodologically sound studies done to date, performed no better than the control

194

KATHLEEN BIEBEL AND JEFFREY L. GELLER

groups, which have mostly been services as usual. Systems of care needs further study to determine in what ways, or for what populations, or for which state departments, or through what manner of integration, or through which levels of government is systems of care clinically, socially, and cost effective. To be an appealing concept in today’s community-embracing, antiinstitutional ethos, a system of care must be an encompassing model, not one that pits community enthusiasts against public hospital advocates. If systems of care are to survive and ﬂourish as a model, this must be understood. To be effective, there must be substantial modiﬁcations of business as usual. State departments and local agencies must shed their historic silo mentalities. Traditional funding streams require major redirections. Children, adolescents, and parents need fundamental re-education about how the mental health system works and how to access its components. Whether or not the efforts are worth making depends on demonstrating the effectiveness of systems of care. But to have even a chance for success, any system of care needs to be adequately funded and adequately staffed with personnel trained to deliver the requisite services to children and adolescents with SED. And the children and adolescents must be those with SED, not a cornucopia of youth that ‘‘no one knows what to do with.’’ The current agenda, therefore, should be the adequate funding, at the state and federal levels, of well-designed studies of systems of care. Parsing the components of systems of care, and either using those that are effective and modifying others, or incorporating the effective components into other models of care and treatment, should be the way to move the system of care agenda. If entire systems of care are demonstrably effective, then this model can become an evidence-based practice endorsed by federal agencies, e.g., CMHS; professional organizations, e.g., APA, AACAP; and state governments, e.g., departments of mental health.

REFERENCES 2003 APA Gold Award. A more compassionate model for treating children with severe mental disturbances. Psychiatric Services, 54, 1529–1531. American Academy of Child and Adolescent Psychiatry. (1989, June). Inpatient hospital treatment of children and adolescents. Retrieved December 29, 2004, from http://www.aacap.org/publications/policy/ps16.htm American Psychiatric Association. (1989, June). Psychiatric hospitalization of children and adolescents. Position statement. Arlington, VA: Author.

Challenges for a System of Care

195

American Psychiatric Association. (2002, January). Addressing the mental health needs of America’s children. Retrieved February 8, 2005, from http://www.psych.org/ advocacy_policy/leg_issues/fac-children.cfm American Psychiatric Association Task Force on Quality Indicators and APA Task Force on Quality Indicators for Children. (2002). Quality indicators. Washington, DC: American Psychiatric Association. Arnold, E. M., Goldston, D. B., Ruggiero, A., Reboussin, B. A., Daniel, S. S., & Hickman, E. A. (2003). Rates and predictors of rehospitalization among formerly hospitalized adolescents. Psychiatric Services, 54, 994–997. Bazelon Center for Mental Health Law. (1999). Making sense of Medicaid for children with serious and emotional disturbance. Washington, DC: Author. Bazelon Center for Mental Health Law. (2004). Press release: Thousands of children with mental illness warehoused in juvenile detention centers awaiting mental health services. Judge David L. Washington, DC: Bazelon Center for Mental Health Law. Bickman, L. (1996). A continuum of care. More is not always better. American Psychologist, 51, 689–701. Bickman, L., Andrade, A. R., & Lambert, E. W. (2002). Dose response in child and adolescent mental health services. Mental Health Services Research, 4, 57–70, 2002. Bickman, L., Foster, M., & Lambert, W. (1996). Who gets hospitalized in a continuum of care? Journal of the American Academy of Child and Adolescent Psychiatry, 35, 74–80. Bickman, L., Lambert, E. W., Andrade, A. R., & Penaloza, R. V. (2000). The Fort Bragg continuum of care for children and adolescents: Mental health outcome over 5 years. Journal of Consulting and Clinical Psychology, 68, 710–716. Bickman, L., Noser, L., & Summerfelt, W. T. (1999). Long-term effects of a system of care on children and adolescents. Journal of Behavioral Health Services and Research, 26, 185–202. Bickman, L., Salzer, M. S., Lambert, E. W., Saunders, R., Summerfelt, W. T., Heﬂinger, C. A., & Hamner, K. (1998). Rejoinder to Mordock’s critique of the Fort Bragg evaluation project: The sample is generalizable and the outcomes are clear. Child Psychiatry and Human Development, 29(1), 77–91. Bickman, L., Summerfelt, T., & Foster, M. (1996). Research on systems of care. In: B. A. Stroul (Ed.), Children’s mental health. Baltimore: Paul H. Brooks. Bickman, L., Summerfelt, W. T., & Noser, K. (1997). Comparative outcomes of emotionally disturbed children and adolescents in a system of services and usual care. Psychiatric Services, 48, 1543–1548. Biebel, K., & Katz-Leavy, J. (2005). Maximizing medicaid options for children with serious emotional disturbances. Unpublished raw data. Bladder, J. C. (2004). Symptoms, family, and service predictors of children’s psychiatric Rehospitalization within one year of discharge. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 440–451. Brinkman-Sull, D. C., Overholser, J. C., & Silverman, E. (2000). Risk of future suicide attempts in adolescent psychiatric inpatients at 18-month follow-up. Suicide and Life-Threatening Behavior, 30, 327–340. California Institute for Mental Health. (2001, August). Psychiatric hospital beds in California: Reduced numbers create system slow-down and potential crisis. A report from the California Institute for Mental Health. California: Author.

196

KATHLEEN BIEBEL AND JEFFREY L. GELLER

Callahan, J. J., & Merrick, E. L. (1997). Designing public sector managed care systems. In: K. Minkoff & D. Pollack (Eds), Managed mental health care in the public sector: A survival manual (pp. 45–58). Amsterdam, Netherlands: Harwood Academic Publishers. Center for Mental Health Services. (1999). Annual report to Congress on the evaluation of the comprehensive community mental health services for children and their families program, 1999. Atlanta, GA: ORC Macro. Challenges facing behavioral health care. (2003). Challenges facing behavioral health care. Washington, DC: National Association of Psychiatric Health Systems. Cocozza, J. J., & Skowyra, K. R. (2000). Youth with mental health disorders: Issues and emerging responses. Juvenile Justice, 7(1), 3–13. Cook, J. R., & Kilmer, R. P. (2004). Evaluating systems of care: Missing links in children’s mental health research. Journal of Community Psychology, 32, 655–674. Cornsweet, C. (1990). A review of research on hospital treatment of children and adolescents. Bulletin of the Menninger Clinic, 54(1), 64–78. Danielson, C. K., Overholser, J. C., & Butt, Z. A. (2003). Association of substance abuse and depression among adolescent psychiatric inpatients. Canadian Journal of Psychiatry, 48, 762–764. DeLeo, D., & Evans, R. (2004). International suicide rates and prevention strategies. Cambridge, MA: Hogrefe & Huber Publishers, Inc. Deutsch, A. (1948). The shame of the states. New York: Harcourt, Brace & Co. Donovan, A., Siegel, L., Zera, G., Plant, R., & Martin, A. (2003). Seclusion and restraint reform: An initiative by a child and adolescent psychiatric hospital. Psychiatric Services, 54, 958–959. Dore, M. M., Wilkinson, A. N., & Sonis, W. A. (1992). Exploring the relationship between a continuum of care and intrusiveness of children’s mental health services. Hospital and Community Psychiatry, 43, 44–48. Evans, M. E. (2003). An experimental study of the effectiveness of intensive in-home crisis services for children and their families: Program outcomes. Journal of Emotional and Behavioral Disorders. Retrieved January 20, 1005, from http://www.ﬁndarticles.com/p/ articles/mi_mOFCB/is_2_11/ai_10312005/print Evans, M. E., & Armstrong, J. I. (2005). A mixed method study of collaboration in systems of care. In: C. Newman, C. Liberton, K. Kutash & R. M. Friedman (Eds), The 17th annual research conference proceedings: A system of care for children’s mental health: Expanding the research base (pp. 19–21). Tampa, FL: University of South Florida, Louis de la Parte Florida Mental Health Institute, Research and Training Center for Children’s Mental Health. Evans, M. E., Boothroyd, R. A., & Armstrong, M. I. (1997). Development and implementation of an experimental study of the effectiveness of intensive in-home crisis services for children and their families. Journal of Emotional and Behavioral Disorders, 5, 93–105. Farmer, E. M. Z., Burns, B. J., Phillips, S. D., Angold, A., & Costello, E. J. (2003). Pathways into and through mental health services for children and adolescents. Psychiatric Services, 54, 60–66. Geller, J. L. (1998). Mental Health Services for the future: Managed care, unmanaged care, mismanaged care. In: G. Schamess & A. Lightburn (Eds), Humane managed care? Washington, DC: NASW. Geller, J. L. (2000). The last half century of psychiatric services as reﬂected in Psychiatric Services. Psychiatric Services, 51, 41–67. Geller, J. L., & Biebel, K. (in press). The premature demise of public child and adolescent inpatient psychiatric beds. Part I: Overview and current conditions. Psychiatric Quarterly.

Challenges for a System of Care

197

Geller, J. L., & Biebel, K. (in press). The premature demise of public child and adolescent inpatient psychiatric beds. Part II: Challenges and implications. Psychiatric Quarterly. Geller, J. L., & Fisher, W. H. (1993). Transitional residences in the linear continuum: Debunking the myth. American Journal of Psychiatry, 150, 1070–1076. Goldstrom, I., Jaiquan, F., Henderson, M., Male, A., & Manderscheid, R. (2001). The availability of mental health services to young people in juvenile justice facilities: A national survey. In: R. W. Manderscheid & M. J. Henderson (Eds), Mental health, United States, 2000. Rockville, MD: U.S. Department of Health and Human Services, Center for Mental Health Services. Gorman, M. (1956). Every other bed. Cleveland, OH: World Publishing Co. Greenbaum, P. E., Dedrick, R. F., Friedman, R. M., Kutash, K., Brown, E. C., Lardieri, P., et al. (1996). National Adolescent and Child Treatment Study (NACTS): Outcomes for children with serious emotional and behavioral disturbance. Journal of Emotional and Behavioral Disorders, 4(3), 130–146. Greenﬁeld, B., Larson, C., Hechtman, L., Rousseau, C., & Platt, R. (2002). A rapid-response outpatient model for reducing hospitalization rates among suicidal adolescents. Psychiatric Services, 53, 1574–1579. Grob, G. N. (1994). The mad among us. Cambridge, MA: Harvard University Press. Harrington, R., Kerfoot, M., Dyer, E., McNiven, F., Gill, J., Harrington, V., et al. (1998). Randomized trial of a home-based family intervention for children who have deliberately poisoned themselves. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 512–518. Healy, D. (2004). Let them eat prozac. New York: New York University. Heﬂinger, C. A., Simpkins, C. G., & Foster, E. M. (2002). Modeling child and adolescent psychiatric hospital utilization: A framework for examining predictors of service use. Children’s Services: Social Policy, Research, and Practice, 5, 151–171. Henggeler, S. W. (1999). Multisystemic therapy: An overview of clinical procedures, outcomes, and policy implications. Child Psychology and Psychiatry Research, 4, 2–10. Henggeler, S. W., Rowland, M. D., Halliday-Boykins, Z., Sheidow, A. J., Ward, D. M., Randall, J., et al. (2003). One-year follow-up of multisystemic therapy as an alternative to the hospitalization of youths in psychiatric crisis. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 543–551. Henggeler, S. W., Rowland, M. D., Randall, J., Ward, D. M., Pickrel, S. G., Cunningham, P. B., et al. (1999). Home-based multisystemic therapy as an alternative to the hospitalization of youths in psychiatric crisis: Clinical outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1331–1339. Hoagwood, K., Jensen, P. S., Petti, T., & Burns, B. J. (1996). Outcome of mental health care for children and adolescents: I. A comprehensive conceptual model. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1055–1063. Hoger, C., Zieger, H., Presting, G., Witte-Lakemann, G., Specht, F., & Rothenberger, A. (2002). Predictors of length of stay in inpatient child and adolescent psychiatry: Failure to validate an evidence-based model. European Child and Adolescent Psychiatry, 11, 281–288. Huey, S. J., Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C., Cunningham, P. B., Pickrel, S. G., et al. (2004). Multisystemic therapy effects on attempted suicide by youths presenting psychiatric emergencies. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 183–190.

198

KATHLEEN BIEBEL AND JEFFREY L. GELLER

Jensen, P. S., Hoagwood, K., & Petti, T. (1996). Outcomes of mental health care for children and adolescents: II. Literature review and application of a comprehensive model. Journal of the American Academy of Child and Adolescent Psychiatry,, 35, 1064–1077. Kanapaux, W. (2004, March). Child psychiatry faces workforce shortage. Psychiatric Times, pp. 1–2, 7. Kaplan, S., Busner, J., Chibnail, J., & Kang, G. (2001). Consumer satisfaction at a child and adolescent stale psychiatric hospitals. Psychiatric Services, 52, 202–206. Kelly, T. M., Donovan, J. E., Cornelius, J. R., Bukstein, O. G., Delbridge, T. R., & Clark, D. B. (2003). Psychiatric disorders among older adolescents treated in emergency departments on weekends: A comparison with a matched community sample. Journal of Studies on Alcohol, 64, 616–622. Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., et al. (1994). Lifetime and 12-month prevalence of DSM-III-F psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry, 51, 819. Kiesler, C. A. (1993). Mental health policy and the psychiatric inpatient care of children. Applied and Preventive Psychology, 2, 91–99. Lourie, I. S., Katz-Levy, J., & Stroul, B. A. (1996). Individualized services in a system of care. In: B. A. Stroul (Ed.), Children’s mental health. Baltimore: Paul H. Brooks. Lyons, J. S., Kisiel, C. L., Dulcan, M., Cohen, R., & Chesler, P. (1997). Crisis assessment and psychiatric hospitalization of children and adolescents in state custody. Journal of Child and Family Studies, 6, 311–320. Manteuffel, B., Stephens, R. L., & Santiago, R. (2002). Overview of the national evaluation of the comprehensive Community Mental Health Services for Children and Their Families Program and summary of current ﬁndings. Children’s Services: Social Policy, Research, and Practice, 5(1), 3–20. Massachusetts Citizens for Children. (2001, April). Treatment and intervention: The essentials to healing. A state call to action: Working to end child abuse and neglect in Massachusetts. Retrieved March 15, 2005, from http://www.masskids.org/cta/ Mayes, S. D., Calhoun, S. L., Krecko, V. F., Vesell, H. P., & Hu, J. (2001). Outcome following child psychiatric hospitalization. Journal of Behavioral Health Services and Research, 28, 96–103. McDermott, B. M., McKelvey, R., Roberts, L., & Davies, L. (2002). Severity of children’s psychopathology and impairment and its relationship to treatment setting. Psychiatric Services, 53, 57–62. Molidor, C. E. (1997). Treatment of children and adolescents. In: T. P. Watkins & J. W. Callicut (Eds), Mental health policy and practice tomorrow. Thousand Oaks, CA: Sage. Mordock, J. B. (1997). The Fort Bragg continuum of care demonstration project: The population served was unique and the outcomes are questionable. Child Psychiatry and Human Development, 27, 241–254. Mordock, J. B. (1998). You can go anywhere from where you are: A response to the rejoinder. Child Psychiatry and Human Development, 29, 93–99. Polnick, E. (1998). Madness on the couch. New York: Simon and Schuster. Pottick, K., Hansell, S., Gutterman, E., & White, H. R. (1995). Factors associated with inpatient and outpatient treatment of children and adolescents with serious mental illness. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 425–433. Ringel, J. S., & Sturm, R. (2001). National estimates of mental health utilization and expenditures for children in 1998. Journal of Behavioral Health Services and Research, 28(3), 319–333.

Challenges for a System of Care

199

Romansky, J. B., Lyons, J. S., Lehner, R. K., & West, C. M. (2003). Factors related to psychiatric hospital readmission among children and adolescents in state custody. Psychiatric Services, 54, 356–362. Rosenblatt, J. A., Rosenblatt, A., & Biggs, E. E. (2000). Criminal behavior and emotional disorder: Comparing youth served by the mental health and juvenile justice system. Journal of Behavioral Health Services and Research, 27(2), 221–237. Scallet, L., Bush, S., Rockwood, M., & Fields, E (2000). Enhancing youth services. Retrieved January 5, 2005, from http://www.naphs.org/youth_services/lewinpaper.html Scallet, L. J. (1996). Managing care for mental illness: Paradox and pitfalls. Managed Care Quarterly, 4(3), 93–99. Scharer, K., & Jones, D. S. (2004). Child psychiatric hospitalization: The last resort. Issues in Mental Health Nursing, 25, 79–101. Schoenwald, S. K., Ward, D. M., Henggeler, S. W., & Rowland, M. D. (2000). Multisystemic therapy versus hospitalization for crisis stabilization of youth: Placement outcomes 4 months post-referral. Mental Health Services Research, 2, 3–12. Sharky, J. (1994). Bedlam. Greed, profiteering, and fraud in a mental health system gone crazy. New York: St. Martin’s Press. Sheidow, A. J., Bradford, W. D., Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C., Schoenwald, S. K., et al. (2004). Treatment costs for youths receiving multisystemic therapy or hospitalization after a psychiatric crisis. Psychiatric Services, 55, 548–554. Shera, W. (1996). Managed care and people with severe mental illness: Challenges and opportunities for social work. Health and Social Work, 21(3), 196–206. Silver, S. E., Duchnowski, A. J., Kutash, K., Friedman, R. M., Eisen, M., Prange, M. E., et al. (1992). A comparison of children with serious emotional disturbance served in residential and school settings. Journal of Child and Family Studies, 1(1), 43–59. Sourander, A., & Piha, J. (1998). Three-year follow-up of child psychiatric inpatient treatment. European Child and Adolescent Psychiatry, 7, 153–162. Strauss, G., Chassin, M., & Lock, J. (1995). Can experts agree on when to hospitalize adolescents? Journal of the American Academy of Child and Adolescent Psychiatry, 34, 418–424. Stroul, B. A., & Friedman, R. M. (1986). A system of care for children and youth with severe emotional disturbances (Rev edn). Washington, DC: Georgetown University Child Development Center, CASSP Technical Assistance Center. Stroul, B. A., McCormack, M., & Zaro, S. M. (1996). Measuring outcomes in systems of care. In: B. A. Stroul (Ed.), Children’s mental health. Baltimore: Paul H. Brooks. Teich, J. L., Buck, J. A., Graver, L., Schroeder, D., & Zheng, D. (2003). Utilization of public mental health services by children with serious emotional disturbances. Administration and Policy in Mental Health, 30, 523–534. Tsesis, A. V. (1998). Protecting children against unnecessary institutionalization. 39 South Texas Law Review 995. United States Department of Health and Human Services. (1999). Mental health: A report of the surgeon general. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health. United States House of Representatives Committee on Government Reform. (2004, July). Incarceration of youth who are waiting for community mental health services in the United States. U.S. House of Representatives, Minority Staff Special Investigations Division. Valenstein, E. S. (1986). Great and desperate cures. New York: Basic Books.

EXCHANGING GLANCES? SYSTEMS, PRACTICE, AND EVIDENCE IN CHILDREN’S MENTAL HEALTH SERVICES Abram Rosenblatt and Laura Compian Two movements dominate contemporary children’s mental health services research: (1) the systems of care philosophy and approach to service delivery (Stroul & Friedman, 1996) and (2) the implementation and development of evidence-based or -supported practices (Chambless & Ollendick, 2001; Kazdin, 2004). Although these two sets of approaches developed independently out of different traditions, methods, and philosophies, they are increasingly becoming interrelated for scientiﬁc, pragmatic, and political purposes. This process promises to improve services to youth with mental health needs and may well determine the ultimate successes and failures of both approaches. Nonetheless, mixing research on systems of care and evidence-based practice remains a complex and challenging task. This chapter delves into the methods by which research on systems of care and research on evidence-based practice can be integrated; on how each tradition can learn from the other; and on how coalescing these approaches can inﬂuence the delivery of mental health services to children and adolescents in community settings. Three key questions will be framed and discussed: (1) What constitutes the intervention? (2) What constitutes evidence? (3) Who is the intended audience for the evidence? Research on Community-Based Mental Health Services for Children and Adolescents Research in Community and Mental Health, Volume 14, 201–237 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0192-0812/doi:10.1016/S0192-0812(06)14010-6

201

202

ABRAM ROSENBLATT AND LAURA COMPIAN

WHAT IS THE INTERVENTION? Systems of care and evidence-based practice possess distinct histories. Though each developed out of attempts to improve services to youth with emotional and behavioral disorders, they did so from perspectives so different as to appear diametrically opposed. Service systems exist at multiple levels, including the practice, program, and system levels (Rosenblatt, 1988, 2005; Rosenblatt & Woodbridge, 2003). Research on health and mental health service systems similarly varies, often by level of the service system, with the research methods, independent and dependent variables, populations of interest, and ultimately the consumers of the research product interacting differentially in the creation and understanding of what constitutes a knowledge base for service delivery. Systems of care and, with limited exceptions, evidence-based practices exist at different levels of the service delivery structure, require and derive from different research approaches, and speak to overlapping but historically different audiences. Systems of Care The system of care approach can be considered, at its core, a systemic, policy-oriented change in the structure and delivery of services (Rosenblatt & Woodbridge, 2003). Systems of care developed alongside the wraparound philosophy, with which it shares considerable commonalities. Both the systems of care approach described initially by Stroul and Friedman (1986) and the wraparound approach (Burchard & Clarke, 1990), associated with pioneering work by Dennis, VanDenBerg, Burchard, Tannen, Lourie, and others, grew initially out of clinical, political, and social movements which drew attention to the troubling plight of youth with severe emotional disturbances. The system of care approach was initially spearheaded through the National Institute of Mental Health (NIMH) and the later Center for Mental Health Services (CMHS) Child and Adolescent Service System Program (CASSP) (Day & Roberts, 1991; National Institute of Mental Health, 1983). Over the past 15 years, the systems of care approach grew in popularity, funding, and resources devoted to its implementation. At the federal level, CMHS began with a relatively small system of care grant implementation initiative in four sites in 1992. Under the current moniker, The Comprehensive Community Mental Health Services Program for Children and their Families, this program has grown dramatically, now funding a total of 121 programs to date across the United States. Individual localities and

Exchanging Glances?

203

states have also implemented funding for the development of service delivery systems based on the systems of care approach. The term ‘‘systems of care’’ is part of the child, and often now adult, mental health lexicon, used both with and without any particular meanings or deﬁnitions in many states and localities to describe service systems and sometimes even programs or practice approaches. The growth in systems of care is remarkable and can only be considered a tremendous success in terms of its popularity across the United States. This growth occurred even in the context of equivocal ﬁndings regarding the effectiveness of these reforms (Rosenblatt, 1998; U.S. Department of Health and Human Services, 1999; Farmer, 2000), a phenomenon about which more will be said later in this chapter. There is, nonetheless, a growing descriptive research based on systems of care, providing information that is shaping policy decisions and service delivery (e.g., Manteuffel, Stephens, & Santiago, 2002; Stroul, Pires, Armstrong, & Zaro, 2002; Brannan, Baughman, Reed, & Katz Leavy, 2002; Holden, De Carolis, & Huff, 2002; Rosenblatt, 1998). By any deﬁnition, a system of care is a complex intervention and this chapter does not attempt to comprehensively deﬁne it (see instead National Institute of Mental Health, 1983; Stroul & Friedman, 1986, 1996). Nonetheless, an understanding of the levels of a system of care is essential to the conceptualization of a system of care. Although human service systems can be analyzed from a wide range of perspectives, current research on systems of care can be thought of as focusing on three levels of analysis: (1) the systems level; (2) the programmatic level; and (3) the clinical level (Rosenblatt, 2005). The Systems Level: Structure, Organization, and Financing Systems of care place a special emphasis on linkages between child serving agencies such as mental health, juvenile justice, social welfare, and education, on community-based care in lieu of restrictive placements, on developing a continuum of services, and, in some cases, on measuring the costs and outcomes of care. All of these elements are also essential to the wraparound process (VanDenberg & Grealish, 1996). Analogously, although wraparound efforts are historically known for their emphasis on individualized services, the unconditional provision of care, and ﬂexible funding, many locales following the systems of care approach incorporate these ideals and services into their service arrays. Both systems of care and the wraparound process fully share the principles of involving parents in all aspects of service delivery. They also emphasize providing culturally competent, community-based, integrated care.

204

ABRAM ROSENBLATT AND LAURA COMPIAN

The Program Level: Components of a System of Care A system of care is composed of programmatic components that can include traditional clinical services such as outpatient and inpatient care or can include more innovative, blended services such as therapeutic foster care, case management, and individualized care. In theory, a system of care should be composed of component services that are themselves effective. However, the overall effectiveness of any of these individual components is not well documented (Kutash & Rivera, 1996). Although some ﬁndings are promising, on the whole, the evidence is either non-existent, sparse, or not encouraging regarding the positive effectiveness of most individual program components. The Clinical Level: Caseworker Behaviors, Skills, and Tools Regardless of the level of innovation at the program or system levels, the ultimate success of any care is dependent on what occurs at the ‘‘clinical’’ level. This level refers to the ways in which a caseworker (or a team of caseworkers) interacts directly with children, their families, and their support systems. Clinical interventions in a system of care may include a range of ofﬁce-based psychotherapeutic approaches such as cognitive–behavioral therapy, family therapy, and play therapy. In a system of care, however, interventions at the clinical level also encompass the interactions of line-level staff with children and families across a continuum of settings and interventions. These settings may include the home, the school, the juvenile courts, and the foster care system. The interventions used in these settings are often loosely articulated and consist of the capacity of line-staff to successfully collaborate with probation ofﬁcers, teachers, and foster parents in providing services to the child and family. Historically, clinical interventions were designed to reside in only one programmatic component of a system of care: outpatient therapy. A similar body of knowledge does not exist regarding more novel treatment approaches such as therapeutically enhanced foster care or individualized service programs. Remarkably little is empirically known about what clinicians actually do when they provide many of the services within more novel care modalities. An exception to this tradition is the work conducted by Henggeler and colleagues on developing multi-systemic therapy (MST; Henggeler & Borduin, 1990; Henggeler, Melton, & Smith, 1992). This treatment approach is consonant with the principles of a system of care and focuses on interventions that are tied to the multiple environments or systems in which a child lives. Considering the range of skills required to work in the natural environments in which children and families live, it is not entirely paradoxical that

Exchanging Glances?

205

the voluminous and relatively sophisticated literature on the outcomes of psychotherapy is often considered inadequate for informing the development of a system of care because of numerous factors, most notably a lack of external validity (Weisz, Huey, & Weersing, 1998; Weisz, Donenberg, Han, & Weiss, 1995a). Research on mental health services to children and adolescents at the clinical level has focused on the efﬁcacy of various treatment interventions for a variety of speciﬁc disorders. Based on a metaanalysis of outcome studies of psychotherapy for children aged 4–18, Weisz, Weiss, Alicke, and Klotz (1987) concluded that the average treated individual was better adjusted after treatment than 79% of those not treated. This cumulative ﬁnding provides initial support for the efficacy of outpatient psychotherapy interventions provided to children presenting with speciﬁc mental health problems and behavioral disorders. In fact, when results from clinic-based trials are examined, the ﬁndings are less positive than those conducted in university-based laboratories (Weisz, Weiss, Han, Granger, & Morton, 1995b). It is at this juncture that the concepts of empirically based and supported practices designed to test the effectiveness of mental health treatments in community settings including systems of care come to prominence. Empirically Supported Treatments Unlike systems of care, empirically based and supported treatments grew out of academic research centers and professional organizations. Empirically supported treatments (ESTs) exist predominantly at the practice level of service delivery. The American Academy of Pediatrics (2005) and the American Academy of Child and Adolescent Psychiatry (n.d.) followed the medical pharmaceutical model (MP; Greewald & Cullen, 1984) utilizing efﬁcacy treatment outcome ﬁndings and the consensus conference model (researchers and practitioners meetings to deliberate best practices) to reach an agreement on practice guidelines and parameters. Striving toward an equally scientiﬁcally sound and rigorous empirical base, the child mental health community sought consensus in similar ways. Research in the mental health ﬁeld has evolved similarly to the MP model. Treatments are ﬁrst developed in a university laboratory to assess the efﬁcacy of the intervention and later implemented in the ﬁeld to assess the effectiveness or public health impact. Using randomized controlled trials (RCTs), researchers work to isolate the utility of a treatment in direct comparison to an alternative treatment approach. RCTs can afford a unique opportunity to determine the efﬁcacy of a treatment, the probability that the

206

ABRAM ROSENBLATT AND LAURA COMPIAN

intervention will generate positive change under ideal circumstances (Weisz & Jensen, 2001). Treatments evaluated as efﬁcacious within the context of an RCT earn the label ‘‘evidence-based practice’’ or EST. Indeed, the use of these terms has grown exponentially and some researchers have noted that these descriptors have sometimes been used haphazardly and inappropriately without the scientiﬁc evidence and support that the terms originally connoted (Hoagwood, Burns, Kiser, Ringelsen, & Schoenwald, 2001). To manage the use of the term EST, the American Psychological Association (APA) Task Force on Psychological Intervention Guidelines and the Division 12 Task Force on Promotion and Dissemination of Psychological Procedures put forth criteria for evaluating the degree of internal and external validities of outcome studies (Task Force on Promotion and Dissemination of Psychological Procedures, 1995; Task Force on Psychological Intervention Guidelines, 1995). In general, the APA Task Force on Psychological Intervention Guidelines highlighted the utility of randomized clinical trials with direct comparison of the treatment in question alongside a valid control group in determining the efﬁcacy of the treatment. Shortly thereafter, the APA Task Force on Promotion and Dissemination of Psychological Procedures provided additional detail regarding necessary and sufﬁcient criteria for deeming a treatment as ‘‘empirically supported’’ (Chorpita, Daleiden, & Weisz, 2005). The APA Clinical Division Task Force followed these developments with recommendations for characterizing ‘‘well-established’’ treatments (requiring at least two randomized clinical trials with active controls) and ‘‘probably efﬁcacious’’ (treatments with at least two randomized clinical trials with waitlist controls, a single randomized clinical trial with an active control, or a small series of single case design experiments using active treatment comparisons) (Chorpita et al., 2005). Unfortunately, none of the above guidelines dealt with the unique developmental issues of youth mental health outcomes research. Thus, the APA Clinical and Child and Adolescent Division developed guidelines aimed at the study of youth mental health outcomes (Lonigan, Elbert, & Johnson, 1998). Complementing child-focused research, Kazdin (1999) suggested additional standards for evaluating child mental health services based on four requirements. The four requirements focus less on speciﬁcs of study design and more on the need for a theory of change and linking the change mechanism with child mental health outcomes. Evidence-based or -supported treatments constitute the backbone of academic research in mental health service delivery, though they traditionally have not been particularly popular methods of service delivery in community settings. This situation is rapidly changing, however, as NIH and other

Exchanging Glances?

207

funding sources began to lay out roadmaps for bridging science, practice, and translational research (National Advisory Mental Health Council, 1999). CMHS has incorporated evidence-based practice into the Children’s Mental Health Initiative, and many states have efforts underway to identify, select, and implement evidence-based or -supported practices. Some evidence-based and -supported practices are being widely implemented, most notably MST (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998) which currently has licensed programs in 34 states. Though MST is, in many ways, atypical if not unique among current evidence-based practices, the popularity of MST is emblematic of the growth of these types of treatments in community settings. Problems in Definitions and Implementation: Opposites do not Necessarily Attract Systems of care and evidence-based or -supported treatments share problems in deﬁnition and implementation, albeit in opposite ways. Given the multi-layered nature of systems of care, assessing whether a system can or should be called a ‘‘system of care’’ is a difﬁcult task. ESTs have their own problems in deﬁnition. Concepts deriving out of program evaluation, such as the strength and integrity of an intervention, can be applied to help understand the dimensions relevant to whether a system of care or an EST follows the intent of the model. In addition, complex social interventions such as systems of care and increasingly EST brought into community settings may evolve and change over time, requiring an ecological approach to understanding the evolution of the care systems. Strength and Integrity In order for a program or treatment to be effective, it must be implemented with sufﬁcient strength and integrity (e.g., Sechrest, West, Phillips, Redner, & Yeaton, 1979; Sechrest & Rosenblatt, 1987). The most straightforward analogy is to medicine: penicillin may cure an infection, but if it is administered in an insufﬁcient dose (strength) or on an irregular schedule (integrity), it will not be effective. Similarly, systems of care may constitute a completely correct theory of service delivery. Nonetheless, if the intensity of services that are provided in a given community is insufﬁcient, if interagency coordination is not complete, if staff commitment is lacking, and if programs within a continuum of care are ineffective, then the system of care may not be strong enough to impact on the children and families served. Likewise, if treatment teams are not composed of the people necessary to

208

ABRAM ROSENBLATT AND LAURA COMPIAN

create change for a child and family, if treatment plans are poorly conceived, if children and families irregularly receive needed services, if target populations are not clearly speciﬁed, and if interagency coordination does not exist, then the program as implemented may not have sufﬁcient integrity to the treatment model to achieve results. The lack of strength and integrity in the implementation of a program can lead to erroneous conclusions regarding whether a program, system, or treatment can be or is effective. Evaluation and the Ecology of Systems of Care Assuring that a system of care has sufﬁcient strength and integrity to create outcomes requires ongoing evaluation of elements of the care system to determine which of them are effective. Historically, studies that evaluate the implementation of a program were called process (after Cronbach, 1962) or formative (Scriven, 1967) evaluations. These kinds of evaluations stood in contrast to outcome (Cronbach, 1962) or summative (Scriven, 1967) evaluations. Over two decades ago, however, Tharp and Gallimore (1982) broke down the distinctions between these two types of evaluation by applying the language of ecology to social programs. Although these concepts were designed to describe the development of programs, they are perhaps even more applicable to the development of systems. Tharp and Gallimore argued that social programs evolve over time through a series of relatively transitory stages until they reach a stable condition. This stable condition is described as ‘‘an association of program elements, organized for and producing a deﬁned social beneﬁt, which will continue to exist, and in which there will not be a replacement by other element types, so long as social values, goals and supporting resources remain constant’’ (Tharp & Gallimore, 1982, p. 43). Taken seriously, the implications of outcomes research to consider the evaluation of systems of care from this ecological framework are profound. Most systems of care are relatively new, subject to resource considerations such as grant funding and changes in federal and state reimbursement procedures, and few could meet conditions for program stability. Put another way, systems that are not in a stable condition are unlikely to have sufﬁcient strength and integrity to achieve results. Problems in deﬁnition, strength, integrity, and the ecology of treatments also apply to evidence-based and -supported practices though in different ways than those for systems of care. The problem of integrity is actually inherent to the MP model that underlies how evidence-supported treatments are to move from research to community settings. Following this model, medical intervention and prevention efforts are typically disseminated in generic form with some modiﬁcations in the approach if research has shown

Exchanging Glances?

209

that a particular group may beneﬁt from a tailored approach (Clark et al., 1998; Rakowski et al., 2003). For instance, Lipkus, Lyna, and Rimer (1999) found that information tailored speciﬁcally for low-income African-American smokers in a community health setting was much more beneﬁcial than general education about smoking cessation. However, the ‘‘gaps’’ between efﬁcacy trial and clinical practice conditions may be greater in the ﬁeld of mental health in comparison to public health due, in part, to the additional challenges of implementing a non-biologically based treatment rather than a simple dosage of a medication (Weisz, Chu, & Polo, 2004). Biologically based treatments can be more easily generated, measured, and administered with the perfect precision of a pipette or a syringe. Rarely does the background of the administrator impact the overall effectiveness of the treatment. Mental health interventions, however, are less accurate due to a conﬂuence of factors including the longer and amorphous nature of the ‘‘dosage’’ of most mental health treatments (e.g., 50 min talk therapy session versus one shot of medication) and the more non-speciﬁc, yet powerful elements of attention and the therapeutic relationship (Jensen, Weersing, Hoagwood, & Goldman, 2005). As an example of issues in deﬁnition and determination of the evidence base in mental health, Weisz et al. (2004) described a recent movement by the National Children’s Law Network who brieﬂy considered taking on the cause of improving the visibility and dissemination of empirically supported youth mental health interventions. However, the group of child-advocate attorneys was quickly dissuaded once they discovered that there were myriad of EST criteria and signiﬁcant disagreement among research- and practice-oriented psychologists over the value of ESTs for wide-spread dissemination. The child-advocate attorneys are not the only group who has become disenchanted with ESTs in child mental health services research (see Newman & Castonguay, 1999). The implementation of many efﬁcacyproven childhood mental health interventions has produced complex results with questionable effectiveness (Nathan, 2004; Weisz & Jensen, 1999). Indeed, ‘‘there are good reasons to suspect that moving treatments from efﬁcacy trials into clinical practice may not invariably improve outcomes beyond those found in usual clinical care’’ (Weisz et al., 2004, p. 303). The essential problem is that research suggests that once an intervention is disseminated to the community, the intended outcomes are often not reached. Dissemination in this case refers to the very speciﬁc distribution of materials associated with an EST, including treatment manuals and training guidelines into the community context (Silverman, Kurtines, & Hoagwood, 2004). With evidence-based treatments, the problems are usually ones of

210

ABRAM ROSENBLATT AND LAURA COMPIAN

integrity of the treatment in the community setting: whether the treatment remains as it was in the research setting. Most evidence-based treatments do have sufﬁcient strength in the research setting to achieve stated outcomes. The problem of treatment integrity is multi-faceted and goes under different names and terminologies in the literature regarding evidence-based treatments. One core component of maintaining the integrity of the treatment is assuring the integrity of the population being served, of the match between the population and the intervention. Signiﬁcant differences often exist between the youth generally treated in ‘‘research therapy’’ in comparison to youth seen in community settings (Weisz, 2000). Youth involved in research therapy are carefully screened prior to enrollment in an efﬁcacy trial in order to generate a ‘‘pure’’ sample similar in regard to diagnosis, age, and other variables of interest. However, relative to clients typically found in the research therapy setting, community youth generally have more symptom impairment (Rosenblatt & Rosenblatt, 2000), more comorbidity (Greenbaum, Prange, Friedman, & Silver, 1991; Rosenblatt, Rosenblatt, & Biggs, 2000), are less likely to stay in treatment (Armbruster & Fallon, 1994), and are more likely to have come from low-income, single-parent homes (Southam-Gerow, Weisz, & Kendall, 2003). Community sample youth also differ from research therapy youth in the circumstances under which the child or adolescent comes to therapy. Many youth in the community become involved with mental health treatment because it is mandated due to a brush with the juvenile justice system or they have not succeeded in the traditional academic environment. Therefore, youth typically found in community settings are less motivated to work in treatment and their family’s investment in treatment is also questionable given that they did not self-refer. In addition, the training and skills of those providing the treatment pose problems for the integrity of the treatment. Clinicians involved in research therapy differ from clinicians in community settings on several key variables (Hennggeler, Schoenwald, Liao, Letourneau, & Edwards, 2002). Generally, in research therapy, highly trained graduate students specializing in psychology administer the treatment. Once the treatment enters the community setting, the level of training practitioners receive in the EST varies. In many instances, community-based clinicians are not able to administer the treatment as originally conceptualized because of lack of access to training or information. Sometimes even intensive training prior to dissemination does not guarantee a successful transfer rate (Sorensen et al., 1988). Previous research has shown that the quality of dissemination is increased

Exchanging Glances?

211

with ongoing consultation and supervision for community-based clinicians (Henggeler et al., 2002; Martin, Herie, Turner, & Cunningham, 1998; Schoenwald, Sheidow, & Letourneau, 2004). The background and training of the clinicians is clearly instrumental in terms of the integrity with which the treatment is disseminated. Other factors within the community that inﬂuence the successful dissemination include a lack of sufﬁcient resources (Schoenwald & Hennggeler, 2004), the organizational structure of the adopting agency (Schoenwald, Sheidow, Letourneau, & Liao, 2003), negative or suspicious attitudes of administrators or clinicians toward the new treatment (Stirman, CritsChristoph, & DeRubeis, 2004), and length of time the program has been funded (Rosenblatt et al., 2001). Some argue, however, that complete ﬁdelity should not be the goal within the context of EST dissemination to the community. Most researchers package efﬁcacious ESTs for transport to the community just as it was conceptualized and tested in research therapy. Indeed, there is an implicit assumption that researchers developing ESTs have taken into account the ﬁt between the treatment and the context of delivery (Hoagwood et al., 2001). However, others suggest that it is important to adapt the approach to ﬁt the unique needs and resources of the community to receive the treatment (Jensen, 2003). Summary Both systems of care and ESTs struggle with problems of integrity and ecology in implementation. Problems of the strength of treatment tend to cluster more in systems of care where linkages between the intervention and outcomes can be more diffuse and are usually not demonstrable in more controlled settings. Systems of care contain complexities which make assessing integrity difﬁcult, and the principles underlying systems of care tend to evolve. Evidence-supported treatments struggle with maintaining sufﬁcient integrity in community settings with respect to the populations served, the training of clinical staff, and the accurate dissemination treatment protocols. As ESTs move to community settings, complexities in those settings are likely to attenuate implementation, creating the need to understand the ecology of these treatments. In particular, there may be delays as to whether the treatment is sufﬁciently evolved and stable in the new setting to achieve outcomes. Interventions need to have sufﬁcient strength and integrity to achieve outcomes and need to be sufﬁciently stable and developed to sustain outcomes over time. Developing such interventions is an essential component to create a knowledge base on the impacts of services delivered

212

ABRAM ROSENBLATT AND LAURA COMPIAN

in community settings. The creation of such interventions, however necessary, is not sufﬁcient as research is needed to establish the evidence for the impacts of these interventions.

WHAT IS EVIDENCE? The research based on systems of care and the research based on empirically supported practices are so different that they defy comparisons. System of care research is, by its very deﬁnition and nature applied, occurring in community settings. Evidence-based and -supported practices develop initially in research and academic settings before moving, if at all, into the community. System of care research virtually precludes randomized designs, whereas such designs are the standard for evidence-based and -supported treatments. Systems of care research often is designed to inﬂuence and inform public policy; research on evidence-based and -supported treatments is, at least initially, designed to inﬂuence and inform the scientiﬁc community. Dependent variables are most often different, the populations being treated are most often different, and those conducting the research most often work in different settings. Strikingly, however, systems of care and ESTs do share one key characteristic: a general dissatisfaction with the existing evidence base for effectiveness in community settings and a lack of consensus on how to remedy gaps in the knowledge base. The two traditions do, however, vary as to the reasons for the lack of convincing evidence. Empirically Supported Treatments With regard to ESTs, a signiﬁcant body of literature suggests promising ﬁndings for youth psychotherapy in general and for speciﬁc evidence-based or -supported treatments (Hoagwood et al., 2001). At least 1,500 clinical trials of youth-oriented interventions (Kazdin, 2000) and 500 studies undertaken within the context of controlled research conditions (Weisz & Jensen, 2001) have yielded compelling results (Weisz, Sandler, Durlak, & Anton, 2005). Thus, within the context of efﬁcacy-driven ‘‘research therapy,’’ participants in the treatment condition generally improve relative to a comparison group. However, there is a relatively small evidence base with few empirical studies on the effective dissemination of such child ESTs in community organizations (Herschell, McNeil, & McNeil, 2004). Little remains known about the ways in which efﬁcacious treatments perform in the community contexts that need the most intervention.

Exchanging Glances?

213

Some argue that the effectiveness of most interventions and adaptation to the communities cannot be adequately assessed with traditional evaluation procedures (Hoagwood, Jensen, Petti, & Burns, 1996). Revised evaluation design strategies are needed that better account for the community context (Bruce, Smith, Miranda, Hoagwood, & Wells, 2002). It is only recently that reliable and valid instruments for assessing childhood psychiatric disorders have been developed (Angold et al., 1995; Angold & Costello, 1995; Gadow & Sprafkin, 1994). These measures, however, only assess clinical symptom acuity and functionality. Weisz (2000) suggested that evaluation of treatment effectiveness should be based on a broader range of outcome criteria than symptom and diagnosis. Community-based indicators (Bruce et al., 2002), consumer perspectives, environments (outcomes within these contextual surrounds), and systems assessing ‘‘interwoven layers of impact’’ (Hoagwood et al., 1996, p. 1055) should be considered. Several prominent researchers have examined the extant literature on efﬁcacy and effectiveness research utilizing multivariate analysis (Weisz et al., 1995a; Weisz et al., 1995b) and summaries of the research literature (Hoagwood, 2005; Hoagwood et al., 2001; Jensen et al., 2005; Nathan, 2004; Weisz & Jensen, 2001). Titles of these commentaries and reviews including phrases, such as ‘‘where to from here?’’ (Ollendick & Davis, 2004) and ‘‘The next generation is long overdue’’ (Jensen, 2003), reﬂect the current sense of agitation regarding the status and direction of the ﬁeld within the research and practice camps. These reviews have provided a context for a growing debate on how to close the gap between research and practice with the ultimate goal of developing and disseminating interventions that work as promised in the ﬁeld. Nathan (2004) solemnly suggested that the ﬁeld would not be able to move forward without resolving four key controversies: questions over the validity of efﬁcacy and effectiveness outcome studies, the impact of common factors and treatment factors in outcomes, commonalities among psychosocial treatments, and the debate over therapy as an art or a science. The salient areas of difﬁculty have become increasingly clear; however, there is still a great deal of uncertainty as to how these outstanding issues may be resolved. One potential solution involves a dramatic change in the MP model so that information regarding treatment development and dissemination ﬂows in a more dynamic way. The proposed evolution utilizes a more ﬂuid, bidirectional model of dissemination in which the ﬂow of information moves more readily between practitioners and researchers (King, Hawe, & Wise, 1998). Several researchers have proposed theories and models advocating the utility of intense collaboration between researchers and practitioners

214

ABRAM ROSENBLATT AND LAURA COMPIAN

from the earliest conception of a treatment (Martell & Hollon, 2001; Weisz, 2000). The deployment-focused model espoused by Weisz, Jensen, and McLeod (2005) is based on the notion that treatment effectiveness can be most easily improved if treatments are shaped to ﬁt the community context and extensive evaluation is conducted on the effectiveness of that implementation. Indeed, there are very few empirical investigations of disseminations of child mental health treatments (Herschell et al., 2004). Hoagwood, Burns, and Weisz (2002) extended this line of work with the Community Intervention Development (CID) model, which explicitly called for the development of treatment protocols within the speciﬁc context in which the intervention would eventually be deployed. Others have advocated for a similar hybrid of efﬁcacy-effectiveness clinics in which understudied issues in child mental health outcomes research could be explored, such as compliance, comorbidity, and process issues (Klein & Smith, 1999) and organizational strategies for supporting the implementation of evidencebased treatment (Glisson & Schoenwald, 2005). Systems of Care With regard to systems of care, a consensus seems to have emerged regarding the fundamental story line: systems of care result in signiﬁcant positive system level changes and youth in systems of care show improvements, but there is a lack of convincing evidence that the systems of care approach itself is responsible for these positive changes (Farmer, 2000; U.S. Department of Health and Human Services, 1999; Rosenblatt, 2005). Though completed over a decade ago and subject to considerable discussion if not outright controversy, the Fort Bragg study still dominates the research literature regarding the evidence base for systems of care. In Fort Bragg, a comprehensive system of care approach (Tolan & Dodge, 2005) was implemented with treatment goals aimed at providing a continuum of services for those in need. However, ﬁndings from the treatment evaluation revealed higher costs and higher volume with no observed beneﬁcial effects in terms of clinical or functional outcomes (Bickman, 1996a, 1996b, 1996c; Bickman et al., 1995). It is noteworthy that access to services was improved, satisfaction with services was improved, and hospitalization stays were reduced (Hoagwood et al., 2001). The scientiﬁc status of system of care effectiveness research has been widely discussed and has largely focused on a clinical perspective (Hoagwood, Hibbs, Brent, & Jensen, 1995; Rosenblatt, 2005; Stroul, 1993; Rosenblatt & Woodbridge, 2003; Farmer, 2000). In this research, the clinical and functional abilities of youth with severe emotional disturbance who receive direct

Exchanging Glances?

215

services are assessed over time using various psychometrically sound measures (Manteuffel et al., 2002; Rosenblatt, 1998). This research also may focus on concrete indicators of functioning in the community such as arrest rates or school performance. Importantly, there is a great deal of system of care research that is not generally available in scientiﬁc outlets such as peerreviewed journals or edited volumes (Rosenblatt, 1998; Farmer, 2000). Typically, researchers have utilized program evaluation and clinical research designs to compare systems of care to traditional service delivery systems. These studies provide rich and varied data sets that have been explored through a variety of analytical techniques (e.g., Bickman, 1996a, 1996b, 1996c; Rosenblatt, 1998; Stroul & Friedman, 1996), consistently demonstrating that the system of care reform process yields systemic changes in service delivery. However, the children’s mental health research ﬁeld currently ﬁnds itself in a quandary described by Farmer (2000), where there is a lack of convincing evidence regarding the effectiveness of these reforms at the individual child and family level. Three Key Components of Health Services Research: Effectiveness, Efficiency, and Equity Conducting research on effectiveness of systems of care is fundamentally complex (Farmer, 2000; Rosenblatt, 1998). Complexity is everywhere: in the nature of the service system reform, the children and families served, and the research questions posed. Such complexity, however, is not unique to the system of care movement. Further, as they enter community settings, ESTs are encountering the same problems with complexity of ‘‘real world’’ research environments. Historically, health services researchers have tackled similar problems, ranging from the study of the effectiveness of managed care to speciﬁc medical procedures. In a prior paper, we (Rosenblatt & Woodbridge, 2003) made the argument that systems of care needed to be considered more as policy than as an intervention and that health services research methods suitable for the analysis of policy initiatives could be applied to systems of care. In many ways, systems of care are judged along many criteria established by traditional intervention and efﬁcacy research and have been found wanting in the arenas of research design and clinical outcomes. The converse appears, in many ways, to be true of ESTs. They have a strong impact on clinical outcomes and solid research designs, but as they move into the community, they appear wanting with regard to policy and community-oriented outcomes. Consequently, revisiting the knowledge typically generated by health

216

ABRAM ROSENBLATT AND LAURA COMPIAN

services research regarding the impacts of systemic, programmatic, or clinical service level reforms is illustrative not only of systems of care, but also of how research can proceed with regard to ESTs. The knowledge base generated by health services research can be grouped into three broad categories: effectiveness, efﬁciency, and equity (Aday, Begley, Lairson, & Slater, 1998). Effectiveness focuses on the beneﬁts to people’s health provided by health care. There are two primary perspectives on effectiveness research: (1) the macro-level focusing on the health status of populations (Evans, Barer, & Marmor, 1994; Milio, 1983) and (2) the microlevel focusing on how the interactions of patients, providers, and institutions impact on the health of those served (Brook & Lohr, 1985; Donabedian, 1982; Wennberg, 1990). The macro-level approaches typically consider how environment, behavior, biology, and medical care interact in determining the health status of a population. The micro-level approaches focus on the delivery of health care and the impact of that delivery on the health status of individuals. The second research domain, efﬁciency, has two key dimensions: productive efﬁciency (producing services at the lowest cost) and allocative efﬁciency (maximizing health given constrained resources). Allocative efﬁciency depends on the relative cost effectiveness of investments in improving health. Although productive efﬁciency is typically not a central goal, most systems of care strive for allocative efﬁciency by, for example, using community-based services instead of costly residential care. The drive for efﬁciency is a powerful force in the creation of public policy and a key factor in the adoption of efﬁcacious treatments in children’s mental health. Concerns regarding the efﬁciency of service delivery can easily derive whether interventions are adopted in communities or service systems, regardless of the strength of evidence for their effectiveness. Finally, equity relates to health disparities and the fairness and effectiveness of procedures for addressing these inequities. At the most fundamental level, equity has to do with fair access to appropriate and effective services. Systems of care for children were created, in part, because of clear inequities in the provision of services to youth with severe emotional disturbance. Key rallying cries in children’s mental health such as Knitzer’s (1982) Unclaimed Children highlighted the ways in which children with severe emotional disturbance were inadequately and inappropriately served. Equity with regard to gender, ethnicity, and age is critical in systems of care. Services are provided to address cultural competence and disparities in service delivery so that diverse, younger, and transitional age youth and their families receive a full continuum of services.

Exchanging Glances?

217

The goals of effectiveness, efﬁciency, and equity can collide and produce contradictory information. A service system may be highly efﬁcient yet ineffective and inequitable. As ESTs move from laboratory to community settings, conﬂicts between these goals at multiple levels are likely to arise. The most obvious conﬂicts relate to the cost of efﬁcacious services. Many of the treatments with the most convincing data regarding efﬁcacy are clearly resource intensive, requiring extensive training and low caseloads when compared to standard practice outpatient care. However, such services may be less costly and more efﬁcient than higher levels of care (e.g., Schoenwald, Borduin, & Henggeler, 1998). Similarly, some efﬁcacious treatments may create serious problems with regard to equity. Such treatments may not be effective or applicable to certain sub-populations of youth, or they may be so costly as to reduce the operational capacity of a service system. Subtle to sophisticated interactions between effectiveness, efﬁciency, and equity must be understood if efﬁcacious services are to succeed in community settings. Equity, Efficiency, and Effectiveness and Levels of the Service System Developing the knowledge base for a policy analysis of systems of care is complicated and multi-faceted since the systems exist across multiple levels, have multiple goals, are mutable by design, and can be judged by a wide range of criteria. Table 1 (from Rosenblatt & Woodbridge, 2003) provides a preliminary framework for this research including the three core elements of health services research (effectiveness, efﬁciency, and equity) with the three Table 1. A Framework for Understanding and Conducting Services Research on Systems of Care and Sample Research/Evaluation Topics. System Goals

Practice Level

Program Level

System Level The effect of standards of care, service provision, and/or funding on outcomes The effect of ﬁscal incentives, service system integration, and/or service mix on costs The effect of program mix and/or ﬁscal incentives on disparities

Effectiveness

The effect of a clinical intervention on outcomes

The effect of program philosophy/culture on outcomes

Efﬁciency

The effect of provider productivity on costs

The effects of stafﬁng choices, provider mix, and/or work hours on costs

Equity

The effect of provider choice and decision making on disparities

The effect of program location and accessibility on disparities

218

ABRAM ROSENBLATT AND LAURA COMPIAN

levels of the human service delivery system (practice, program, and system). The table also provides examples of potential areas of focus for research and evaluation efforts when the three levels of service delivery are examined within the three primary domains of health services research. ESTs constitute effectiveness research at the practice level. Clinical interventions such as MST have been tested within real world clinical settings across several different service systems (Henggeler & Borduin, 1990; Henggeler et al., 1992). There are also a number of examples derived from the program evaluation literature that focus on effectiveness at the programmatic level (see Kutash & Rivera, 1996 for a more complete discussion), such as the effects of organizational culture and climate on children’s outcomes (Glisson & Hemmelgarn, 1998). Finally, effectiveness research also exists at the system level; however, the links between service system change and many existing indicators of effectiveness (such as measures of clinical and functional status) are remote. This discussion has been at the core of the current debate regarding the effectiveness of systems of care (Bickman, Noser, & Summerfelt, 1999; Hoagwood et al., 1995, 1996; Rosenblatt, 1998). The efﬁciency of a service system can be measured across the practice and program levels, yet system level reform may have the most direct impact on efﬁciency. For example, the use of restrictive levels of care may be discouraged through systemic emphases on interagency collaboration, the creation of community-based alternatives, and ﬁscal disincentives to residential placement. Managed care initiatives that emphasize capitation create the need for allocative efﬁciency to maximize health beneﬁts within constrained resources. Even without capitation, most children’s mental health systems are forced to engage in various attempts at allocative efﬁciency given constraints on the public and private fundings of mental health services. Although systemic interventions are most often associated with producing efﬁcient services, programmatic and practice level interventions have also been shown to effect efﬁciency. For example, to the degree that an intervention such as MST reduces expensive psychiatric hospital visits while maintaining effectiveness at a lower overall cost (Schoenwald et al., 1998), the intervention may be more efﬁcient than hospitalization. Equity or disparities in the delivery of children’s mental health services can also be addressed through interventions occurring across all levels of the children’s service delivery system. At the practice level, provider choice and decision making regarding eligibility for services may determine whether services are delivered equitably. Similarly, the nature, philosophy, location, and characteristics of a program may determine the equity of a service system. For example, in many service systems, ethnic-speciﬁc programs are

Exchanging Glances?

219

located close to where people of that ethnicity reside. Systemic issues may also broadly inﬂuence equity: access to children’s services can be predicated on eligibility for various funding mechanisms such as Medicaid or private insurance. Disallowing, for example, mental health coverage for migrant workers will affect whether these workers and their children receive services. Effectiveness: The Question of Measurement Although questions of efﬁciency and equity ought to be more thoroughly considered with regard to research on systems of care, and especially with regard to research on ESTs, effectiveness remains the core determinant of whether a treatment is considered empirically supported, and as to whether systems interventions ‘‘work.’’ The question of how to determine the psychometrics of individual measures is thoroughly addressed in numerous texts and journal articles. What is less well addressed, however, is whether most measures used to determine the empirical support for a treatment both work in community settings as expected (see, for example, Rosenblatt & Rosenblatt, 2000) and, more basically, constitute meaningful outcomes in community settings. The question of which outcomes to measure is the one that researchers delving into systems of care frequently address (Rosenblatt, 2005). It is a question, however, that is only now being more extensively considered by researchers who create and test ESTs. The problem of what to measure and the meaning of measures not only is central to the future success of empirically based treatments, but existing problems in measurement can call into question much of the extent literature that underpins the determination of a treatment as empirically based. Kazdin (2006) illustrates how a great deal of the research underlying empirically based treatments is based on ‘‘arbitrary measures’’ (Blanton & Jaccard, 2006). The concept of arbitrary metrics is simple at face value, yet complex and cannot be fully delineated here; however, the core problem is that evidence-based treatments with effects that utilize arbitrary metrics may not actually improve people’s lives. Measures reﬂect arbitrary metrics if the connection between the score on a measure and the true score on the underlying dimension is not known. Traditional statistical assessment of reliability and validity does not address the issue of whether a measure is arbitrary. Though this is a complex conceptual and intellectual problem that may end up occupying psychometricians for some time to come, there are immediate pragmatic implications for ESTs and for systems of care, some of which already appear in other variants in the extant literature. Essentially, most measures of symptoms and of core constructs such as depression, life satisfaction, anxiety, self-esteem, and the like are potentially arbitrary

220

ABRAM ROSENBLATT AND LAURA COMPIAN

metrics unless connected to real world referents. Measures such as arrests, school attendance and progression, mortality, and living situation are not arbitrary. In short, most measures of psychological constructs may well be arbitrary and consequently suspect with regard to real world impacts. This notion has been less directly addressed in some of the literature pertaining to systems of care. We have argued that measures such as whether a youth is ‘‘in home, in school, and out of trouble’’ (Rosenblatt, 1993) should constitute the starting point for understanding the outcomes of systems of care; such measures are indeed less likely to be arbitrary. The concept of arbitrary metrics casts a different light on the multi-dimensionality of outcomes in both systems of care and research on ESTs. For example, our earlier model of the dimensions of outcomes relevant to mental health services research consists of four domains (Rosenblatt & Attkisson, 1997; Rosenblatt, 2005), each of which is inherently more or less likely to contain arbitrary metrics. The ﬁrst domain in this model for outcome measurement is called the clinical status domain. This domain encompasses the dual concepts of mental and physical health and includes measurement and indices of psychopathology and symptomatology encompassing both classiﬁcation and severity. This domain, the one that is most frequently measured in outcome studies, is also the one that is potentially based on arbitrary metrics. The second domain, called functional status, captures the ability to fulﬁll effectively social and role related functions in a variety of life settings. Examples of functional adaptation include the ability to work, attend school, learn, remain in home or maintain independent living, and maintain positive and enhancing social relationships. The metrics in this domain that relate to actual life functioning are not based on arbitrary metrics. The third dimension is called life satisfaction and fulﬁllment. This domain, almost entirely based on arbitrary metrics, relies on prior work that deﬁnes the meaning of terms such as ‘‘well being,’’ ‘‘life satisfaction,’’ ‘‘objective quality of life,’’ ‘‘subjective quality of life,’’ and ‘‘happiness.’’ The life satisfaction and fulﬁllment domain focuses on the subjective appraisal of well being. The ﬁnal domain, welfare and safety, encompasses the safety and welfare problems posed by mental disorders to the individual, their family, their social network, and the community in which they live. Such problems include: selfinjurious behaviors and acts such as suicide, substance abuse, and lack of basic sanitation; infectious diseases such as AIDS, other sexually transmitted disease, and tuberculosis; abuse, neglect, and other forms of violence suffered by the youth; and violent and illegal acts committed by the youth. This domain is composed of indicators that are not likely to be based on arbitrary metrics.

Exchanging Glances?

221

Summary Equity, efﬁciency, and effectiveness can be considered the core information necessary for the development and support of both service system reforms and novel treatments in community settings. Effectiveness is a multidimensional construct, and both empirically based treatments and systems of care tend to emphasize differential types of effectiveness related outcomes. Some outcomes tied to psychological constructs may be based on arbitrary metrics, which may not link directly to real world changes in an individual’s life. Both systems of care and empirically based treatments face challenges with regard to metrics arbitrary or otherwise, and both types of reforms struggle with assessing equity and efﬁciency. Though arbitrary metrics and effectiveness data may sway scientiﬁc audiences, other audiences may be more skeptical of what constitutes convincing evidence for sustaining and implementing novel service approaches.

WHO IS THE AUDIENCE? There are three general purposes in collecting evidence, regardless of the type: (1) to reﬁne and better develop services (e.g., Hernandez & Hodges, 2001); (2) to convince people such as policy makers that the services are effective and should remain funded (e.g., Rosenblatt & Woodbridge, 2003); and (3) to contribute to the generation of scientiﬁc knowledge. Outcome information may be used for various combinations of these purposes to varying degrees. However, it is essential to consider the primary purposes of collecting outcome data, as it can be difﬁcult to meet all potential purposes and satisfy all audiences. The type of measures, methods, and analyses used may vary depending on the type of audience. This does not necessarily mean that any one type of audience will demand lower or higher quality information. Rather, the emphasis as to what is most important in the data collected will vary by audience. Problems often ensue when there is a lack of clarity regarding the purposes of collecting outcome data. Refining Service Delivery Considerable progress has been made in understanding how to utilize outcome information for the development of systems of care for youth with severe emotional disturbance. As described in Hernandez and Hodges (2001), models exist for using theory-based accountability, family empowerment

222

ABRAM ROSENBLATT AND LAURA COMPIAN

models, and performance-based contracting among other strategies to improve service delivery. A range of case studies from California to Vermont to Florida illustrate the power of using outcome information to systematically evolve and improve systems of care for youth with severe emotional disturbance. Schoenwald and Hennggeler (2004) also point to the utility of a continuous quality improvement process in the implementation of ESTs in community settings. The primary audiences for such efforts are children and families, practitioners and service providers, and program administrators. Children and families who are in need of, or who receive, services from the care system are often invested in a range of outcome information. Satisfaction or dissatisfaction with services can, when voiced by consumers and consumer advocates, be a powerful agent for political change. However, consumers are increasingly involved in a range of levels of system development, including program planning and policy. Consequently, although consumers are primarily concerned with the fate of the individual receiving services, they are also often concerned about understanding and overcoming the political and programmatic barriers to improving services and the resources for funding these services. Practitioners often must work at the symptom level, attempting to reduce the occurrence of more harmful thoughts and behaviors and increase the occurrence of more beneﬁcial ways of acting in the world. Clinicians, especially those working within a strength-based model, are likely to be especially concerned about the ability of their clients to remain in school, stay out of trouble, or remain in the home (Rosenblatt, 1993). They are also likely to beneﬁt from clinically oriented measures and other assessment devices that can help them understand the clinical proﬁle and functional strengths and impairments of the child and family. Administrators are likely to ﬁnd functional status data valuable. Information, for example, on work and school performance can point to speciﬁc academic programs that need to be better integrated with the care system. Finally, administrators are often concerned about the cost of services, and so are likely to be focused on reducing utilization of costly and restrictive levels of institutional care. Convincing Stakeholders of the Value of Systems of Care and Empirically Supported Treatments Outcome data are often collected in the hope of convincing policy makers and those in a position to inﬂuence policy makers, to implement or continue a service delivery strategy. Effectiveness information may not be the most

Exchanging Glances?

223

convincing types of information for policy makers. Policy makers might be equally, if not more, convinced by data on questions on efﬁciency (such as producing as many services as possible at a given cost) or on equity (distributing services fairly to those who have need) (Rosenblatt & Woodbridge, 2003). It is quite possible for high quality scientiﬁc information to have limited use to policy makers. For example, an extremely well-controlled study on psychotherapy may have no real relevance to policy makers if the treatment circumstances are not found in the public systems that are their domains. Conversely, policy decisions fueled by factors disregarding sound data may fail as a result of unreasonable, impractical, or hierarchical implementation processes. Policy development that creates and sustains effective service delivery systems must take into account how multiple variables align with system values and resources including: the nature of the problem and population of concern, the ﬁnancing and management of systems, and the organizational operations (Friedman, 1999). These factors require data from various system levels and across multiple stakeholder perspectives. Doing justice to the different models of policy analysis that provide varying perspectives on information found useful to policy makers is far beyond the scope of this chapter (see instead Etzioni, 1967; Hayes, 1992; Lasswell, 1951). Not all types of research are equally likely to impact on the policy process. Aday et al. (1998) suggest three criteria for judging the value of information produced for policy decisions: reasonableness, validity, and relevance. Reasonableness refers to the degree that an argument has internal and external logic, validity refers to the legitimacy and soundness of the research, and relevance refers to the degree of applicability of the research for policy makers. The Value of Outcome Information: Reasonableness, Validity, and Relevance The criteria of reasonableness, validity, and relevance apply to not only how policy makers value outcome data, but also how administrators, practitioners, family members, and scientists value information regarding outcomes. These three criteria for judging the value of research do, however, vary by the purpose of the research. The scientiﬁc community, for example, historically places an emphasis on the validity of the research broadly deﬁned. Until recently, much of the children’s mental health research treatment research placed an emphasis on the internal validity of the research by attending to methods, designs, and measures more than whether the results could be generalized.

224

Table 2.

ABRAM ROSENBLATT AND LAURA COMPIAN

Purposes of Collecting Outcome Data and Criteria for Judging Value.

Purpose

Internal Validity

External Validity

Reasonableness

Relevance

Service reﬁnement Convincing others of value Scientiﬁc knowledge

Low Low High

Medium Medium Medium

High High Low

High High Low

Research designed to improve service delivery must place an emphasis not only on external validity, but also on reasonableness and relevance. Administrators reviewing outcome information are going to utilize their knowledge of the service system in interpreting that information and are going to be particularly concerned with whether the information addresses a question relevant to their decision making process. Using outcome information to convince people of the value of an intervention is going to require an emphasis on all three criteria, with relevance and external validity likely carrying particular weight. Table 2 provides an overview of the likely relative strength of each of these criteria for three different purposes in collecting outcome information. This table illustrates those dimensions that are relatively less important to the purpose of system development or policy formulation and are more likely to guide the creation of information designed primarily to generate scientiﬁc knowledge. Value of Outcomes Research: Dimensions The value of outcomes research for different purposes also varies based on the type of information collected. Outcomes research on systems of care is, by deﬁnition, a multi-dimensional endeavor. The service systems being studied are multi-dimensional and multi-layered, the outcomes being assessed are multi-dimensional and vary by perspective and context, the children and families who receive services are varied in terms of their needs and available resources, and the methods used vary considerably with respect to emphasis and focus. In order to understand how research on systems of care can serve different purposes, it is essential to understand these different dimensions and how they vary if outcome information is to be useful. Levels of the Care System and Outcome Research Goals The purposes of child and family outcomes research are likely to vary by the level of the service system. Research on child and family outcomes is most removed from systemic level interventions. Altering the structure,

Exchanging Glances?

225

organization, and ﬁnance of a service delivery system may impact on child and family outcomes; however, such impacts are usually indirect and ﬁltered through programmatic or practice level interventions. Systemic level interventions are, however, most closely linked to goals of policy formulation. Systemic changes are generally more easily legislated or mandated than changes in program and practice. Local, state, and federal decision makers frequently alter ﬁnancial and reimbursement mechanisms, for example, to meet policy goals. Similarly, state and local mental health commissioners or directors often make systemic alterations in service delivery and are less frequently involved at the programmatic and especially practice levels. At the other end of the spectrum, research at the practice level has the most direct impact on child and family outcomes. Traditionally, the children’s mental health research community has focused most intensively at this level. Most psychotherapy research focuses on practice level interventions, for example. Line-level practitioners are most likely to attend to research at this level, though the limited impacts of research on practice patterns are well documented. Although research at the practice level often contributes to scientiﬁc knowledge, it is infrequently the basis of policy or even practice in real world settings. Practice level research may, however, be used for system accountability purposes. Administrators may track outcomes by clinician, practitioners may keep records of the outcomes of the children and families they serve, and practice patterns may be consequently reviewed and altered. Programmatic level interventions exist between the practice and systemic levels and consequently can serve the purposes of policy formulation, system accountability, and scientiﬁc knowledge creation in relatively equal amounts. It is, for example, quite common for programs to be put in place by legislative or administrative actions. The programs may be evaluated and if outcomes are positive, they may be perpetuated. Similarly, programs may be created, altered, or disbanded based on outcome information by managers of local service systems. There is a growing interest in effectiveness research at the practice level, where empirically based treatment models are adopted and then tested in community-based clinical settings. Clinical interventions such as MST have been tested within real world clinical settings across several different service systems (Henggeler & Borduin, 1990; Henggeler et al., 1992). There are also a number of examples derived from the program evaluation literature that focus on effectiveness at the programmatic level (see Kutash & Rivera, 1996 for a more complete discussion), such as the effects of organizational culture and climate on children’s outcomes (Glisson & Hemmelgarn, 1998). Finally,

226

ABRAM ROSENBLATT AND LAURA COMPIAN

effectiveness research also exists at the system level; however, the links between service system change and many existing indicators of effectiveness (such as measures of clinical and functional status) are remote. This discussion has been at the core of the current debate regarding the effectiveness of systems of care (Bickman et al., 1999; Hoagwood et al., 1995, 1996; Rosenblatt, 1998). The essential point is that although each level of the service system can produce outcomes suitable for internal accountability, policy making, and scientiﬁc knowledge, the natural impacts of research on each level of the service system vary. Consequently, special and unusual steps need to be taken if research at the practice level is to impact on practice just as special and unusual steps need to be taken if research at the system level is to contribute to scientiﬁc knowledge. Practice level research is often produced with the goal of publishing ﬁndings and may never be viewed by decision makers, whereas system level research is often conducted with the goal of being reported to decision makers and may never be published. Table 3 illustrates the likely matches between each level of the service system and the purposes of collecting outcome information. The Purposes of Outcomes Research Revisited The purposes of conducting child and family research, the levels of the service system, and the domains of data collection all intersect. Research on systems of care is not only complex to conduct (Rosenblatt, 1998), but is also a multi-dimensional construct in and of itself. There is no single or dominant type of outcome research when it comes to systems of care for youth with severe emotional disturbance. This contributes to the confusion that often exists when outcome studies are designed and conducted as well as when outcome studies are utilized by a range of stakeholders. Research on ESTs that began in the research setting focused on a scientiﬁc audience; as they enter the community and public arenas, the audience will likely change. Now that the levels of the service system, the criteria for judging value, and the dimensions of measurement have been discussed, it is possible to Table 3. Likely Impacts of Research on Three Levels of the Service System.

Policy formulation Service reﬁnement Scientiﬁc knowledge

System

Program

Practice

High High Low

High High High

Low Medium High

Exchanging Glances?

Table 4.

227

Goals of Outcomes Research and Value, Dimensions, and Service System Level. Service System Level Emphasis

How Value is Primarily Judged

Service reﬁnement

System Program

Relevance Reasonableness External validity

Policy formulation

System Program Program Practice

Relevance Reasonableness Internal validity

Scientiﬁc knowledge

Primary Measurement Domains Clinical status Functional status Satisfaction Safety and welfare Safety and welfare Functional status Clinical status Functional status

revisit the three primary goals of outcomes research. As Table 4 illustrates, each type of outcome research varies in its relative emphasis with regard to value, measurement dimensions, and level of the service system. Research designed to reﬁne service delivery is likely to emphasize system and programmatic level analyses. The value of the information is likely to be judged based on all three criteria, with relatively little less attention to the internal validity. All four domains for measurement are likely to be used. Research designed to guide policy formulation will also likely emphasize systemic and programmatic levels with attention to relevance and reasonableness. Safety and welfare and functional status are the primary measurement domains likely to inﬂuence policy formulation. Research designed primarily to contribute to the scientiﬁc knowledge base emphasizes the program and practice levels, attends highly to internal validity, and typically features measures of clinical and functional status. Of course, none of the information in Table 4 is absolute, policy makers may attend to internal validity and scientists may collect information on safety and welfare. Rather, Table 4 is meant to illustrate the fact that there tend to be differences in relative emphasis. Summary The purpose of this section is to illustrate that there are at least three different cultures of outcomes research organized around the purposes and goals of the research. Not surprisingly, each culture is occupied by varying types of investigators. Policy research is often conducted either by state or local evaluation staff or through contracts with universities or, more frequently, research institutes. Local evaluators working within the service

228

ABRAM ROSENBLATT AND LAURA COMPIAN

system primarily conduct research designed to reﬁne service delivery. Academics are the most frequent contributors to the scientiﬁc knowledge base. Systems of care research, existing virtually entirely in community settings, needs to consider the goals of policy formulation more extensively than research on ESTs which need, at least initially, to meet criteria for scientiﬁc knowledge generation above all else. It is, of course, the case that many individuals cross these boundaries. However, just as there are cultures that need to be bridged in the creation of integrated service delivery systems, so too are there cultures that need to be bridged in the creation of research on integrated service delivery systems. In many ways, the research community pertaining to systems of care is an uncommon mix of academicians, local evaluators, state level researchers, independent consultants, program and service system administrators, and policy makers. Nonetheless, part of the challenge in moving the applied knowledge base forward is that the methods, values, and levels of the service systems vary depending on the research tradition. The consequences of such differences are many. Systemic interventions have been expected to improve individual outcomes without clear causal links to alterations in clinical practice. Similarly, clinical practice interventions that may not alter the characteristics of a service system have been expected to do so.

LINKING INTERVENTIONS, EVIDENCE, AND AUDIENCES The concepts and models articulated in this chapter derive out of both the systems of care and the EST literature. Though presented separately, the myriad of concepts related to deﬁnitions, evidence, and audience presented are indeed linked. The nature and characteristics of the intervention, the evidence collected in support of the intervention, and the audiences for the evidence are best considered in total. Considerable problems can occur when these dimensions are not aligned, when, for example, a primary policy and system level intervention like the system of care approach is treated primarily as a clinical intervention, or, conversely, when interventions designed to improve clinical care and contribute to scientiﬁc knowledge enter both the community and the political arenas without consideration as to whether such a move is warranted or whether the intervention is ready for community-based implementation. As described in this chapter, considerably careful thought is currently going into how one can transport, disseminate, and if necessary adapt

Exchanging Glances?

229

evidence-based treatments into community settings. Similarly, considerable thought, largely in the wake of the Fort Bragg study, has gone into how to assess the impacts of systems of care and whether and in what ways they ‘‘work.’’ When considered together, the concepts from both lines of investigation constitute a fairly extensive, if not comprehensive, set of guideposts regarding research on the implementation of mental health services in community settings. Up to this point, ESTs and systems of care have been primarily discussed apart from one another. However, the key points deriving out of both sets of literature apply equally to ESTs that are destined for community-based settings as well as to systems of care. When conceptualizing, designing, or evaluating research protocols for community-based interventions, the links and interrelationships between the intervention, the nature of the evidence, and the eventual audiences for the evidence are best considered. No single research design or intervention can, of course, be all things to all audiences; however, careful consideration of the components of each of these three primary arenas for service delivery can help avoid potentially problematic mismatches and ultimately unsuccessful efforts at implementation. Interventions can be characterized by the level of the service system in which they reside, either the practice, program, or system levels. Occasionally interventions may exist across multiple levels. They can also be assessed with regard to their strength, whether they can achieve positive results, and their integrity, whether they implemented as intended. The level at which the program resides determines, in part, the kinds of evidence most appropriate to the intervention. Practice level reforms are most likely to impact on effectiveness related outcomes, particularly clinical and functional status. Program level reforms are often hybrids that may impact on multiple outcomes including particularly effectiveness and efﬁciency related variables and the outcomes may span multiple dimensions. System level reforms are most likely to impact on effectiveness outcomes related to safety and welfare, as well as having the highest probability of impacting on efﬁciency and equity. Audiences and goals for different levels also vary, system reforms are the most natural ﬁt for policy makers, practice reforms are the most natural ﬁt for scientiﬁc knowledge development, and programs again exist somewhere between the two. A Simplified Model It is possible to characterize the concepts in this chapter into a uniform model that describes the relationships between interventions, evidence, and

230

ABRAM ROSENBLATT AND LAURA COMPIAN

audiences for both systems of care research and ESTs moving into community-based settings. The most simpliﬁed version of the model initially considered for system of care impact looked like (1): O ¼ f ðSÞ

(1)

In this conceptualization, outcome is a function of system change and outcome refers to clinical outcomes or effectiveness. The most simpliﬁed version of the model driving ESTs is expressed in (2): O ¼ f ðCÞ

(2)

In this conceptualization, outcome is a function of clinical or practice level reforms. In 1998, Rosenblatt proposed a model that was somewhat a combination of the two above expressed in (3): Oitdcp ¼ f ðS; P; CÞ

(3)

where P reﬂects some constellation of programmatic reforms and C some constellation of clinical reforms. Systemic, programmatic, and clinical change may interact with one another within this framework to produce different types of outcomes. This perspective also allows for the exploration of how speciﬁc types of outcomes may only relate to speciﬁc combinations of system, program, and clinical interventions. However, this conceptualization does not fully reﬂect the lessons derived from both systems of care and EST research. The ultimate impact, I, of an intervention can be described by (4): I ¼ f ðS; P; C; A; Oeee Þ

(4)

where impact (I) is a function of systemic, program, and clinical level changes, and the audience (A) and the outcome (O) are related to effectiveness, efﬁciency, and equity. The direct implications of this model for analyzing the outcomes for interventions research are threefold: (1) interventions need to be created, understood, studied, and described, and their strength and integrity assessed, at multiple levels; (2) evidence needs to consider effectiveness, efﬁciency, and equity as well as the audience; and (3) the relationships between levels of intervention, audiences, and evidence can be interactive as well as additive. With some exceptions, the literature on systems of care and the literature and ESTs exist in separate worlds. It is particularly atypical to ﬁnd articles on ESTs that refer to systems of care. Yet, as illustrated in this chapter, these two approaches share common problems and challenges, albeit in

Exchanging Glances?

231

different ways. Each tradition poses lessons for the other, and each is increasingly likely to co-exist if systems of care retains its current popularity in community settings and if more treatments move from research to community environments. In most regards, these two sets of interventions are complementary, though there can be clashes in values and approaches. The strength-based values inherent in systems of care, for example, may not match highly clinically oriented approaches that focus on identifying symptoms and deﬁcits in functioning. Some promising clinical interventions may require well-organized and integrated systems to provide youth who require services, and some changes at the system level may enhance or detract from the implementation of some ESTs. However, integrating research traditions can have positive impacts on the knowledge base. There are few good reasons for not integrating these research approaches, though that, one may assume, is like most things in the world of research, an empirical question.

REFERENCES Aday, L. A., Begley, C. E., Lairson, D. R., & Slater, C. H. (1998). Evaluating the healthcare system: Effectiveness, efficiency and equity. Chicago, IL: Health Administration Press. American Academy of Child and Adolescent Psychiatry. (n.d.). Practice parameters. Retrieved October 11, 2005, from American Academy of Child and Adolescent Psychiatry website www.aacap.org/clinical/parameters/index.htm American Academy of Pediatrics. (2005). Pediatric clinical practice guidelines and policies (5th ed.). Elk Grove Village, IL: American Academy of Pediatrics. Angold, A., & Costello, E. J. (1995). A test–retest reliability study of child-reported symptoms and diagnoses using the child and adolescent psychopathology assessment (CAPA-C). Psychological Medicine, 25, 755–762. Angold, A., Prendergast, M., Cox, A., Harrington, R., Simonoff, E., & Rutter, M. (1995). The child and adolescent psychopathology assessment (CAPA). Psychological Medicine, 25, 739–753. Armbruster, P., & Fallon, T. (1994). Clinical, sociodemographic, and systems risk factors for attrition in a children’s mental health clinic. American Journal of Orthopsychiatry, 64, 577–585. Bickman, L. (1996a). The evaluation of a children’s mental health managed care demonstration. The Journal of Mental Health Administration, 23, 7–15. Bickman, L. (1996b). Reinterpreting the Fort Bragg evaluation ﬁndings: The message does not change. The Journal of Mental Health Administration, 23, 137–145. Bickman, L. (1996c). A continuum of care: More is not always better. American Psychologist, 51(7), 689–701. Bickman, L., Guthrie, P. R., Foster, E. M., Lambert, E. W., Summerfelt, W. T., Breda, C. S., & Heﬂinger, C. A. (1995). Evaluating managed mental health services: The Fort Bragg experiment. New York: Plenum Press. Bickman, L., Noser, K., & Summerfelt, W. T. (1999). Long-term effects of a system of care on children and adolescents. The Journal of Behavioral Health Services and Research, 26, 185–202.

232

ABRAM ROSENBLATT AND LAURA COMPIAN

Blanton, H., & Jaccard, J. (2006). Arbitrary metrics in psychology. American Psychologist, 61(1), 27–41. Brannan, A. M., Baughman, L. N., Reed, E. D., & Katz Leavy, J. (2002). System-of-care assessment: Cross-site comparison of ﬁndings. Children’s services. Social Policy, Research, and Practice, 5, 37–56. Brook, R., & Lohr, K. (1985). Efﬁcacy, effectiveness, variations, and quality: Boundarycrossing research. Medical Care, 23, 710–722. Bruce, M. L., Smith, W., Miranda, J., Hoagwood, K., & Wells, K. B. (2002). Community-based interventions. Mental Health Services Research, 4(4), 205–214. Burchard, J. D., & Clarke, R. T. (1990). The role of individualized care in a service delivery system for children and adolescents with severely maladjusted behavior. Journal of Mental Health Administration, 17(1), 48–60. Chambless, D. L., & Ollendick, T. H. (2001). Empirically supported psychological interventions: Controversies and evidence. Annual Review of Psychology, 52, 685–716. Chorpita, B. F., Daleiden, E. L., & Weisz, J. R. (2005). Identifying and selecting the common elements of evidence based interventions: A distillation and matching model. Mental Health Services Research, 7(1), 5–20. Clark, M. A., Rakowski, W., Ehrich, B., Pearlman, D. N., Goldstein, M., Dube, C. E., Rimer, B. K., & Woolverton, H. (1998). Stages of adopting regular screening mammography: Do women differ in decisional balance within stages? Journal of Health Psychology, 3(4), 491–506. Cronbach, L. J. (1962). Psychological issues pertinent to recent American curriculum reforms. In: G. Nielson (Ed.), Proceedings of the XIV international congress of applied psychology. Vol. 3. Child and education (pp. 139–152). Oxford, UK: Munksgaard. Day, C., & Roberts, M. C. (1991). Activities of the child and adolescent service system program for improving mental health services for children and families. Journal of Clinical Child Psychology, 20(4), 340–350. Donabedian, A. (1982). Explorations in quality assessment and monitoring. volume II – The criteria and standards of quality. Chicago: Health Administration Press. Etzioni, A. (1967). Mixed scanning: A third approach to decision making. Public Administration Review, 27, 385–392. Evans, R. G., Barer, M. L., & Marmor, T. R. (1994). Why are some people healthy and others not? The determinants of health of populations. New York: Aldine De Gruyter. Farmer, E. M. Z. (2000). Issues confronting effective services in systems of care. Children and Youth Services Review, 22, 627–650. Friedman, R. M. (1999). A conceptual framework for developing and implementing effective policy in children’s mental health. Tampa, FL: Research and Training Center for Children’s Mental Health, Louis de la Parte Florida Mental Health Institute, University of South Florida. Gadow, K., & Sprafkin, J. (1994). Child symptom inventories manual. Stoney Brook, NY: Checkmate Plus Limited. Glisson, C., & Hemmelgarn, A. (1998). The effects of organizational climate and interorganizational coordination on the quality and outcomes of children’s service systems. Child Abuse and Neglect, 22, 401–421. Glisson, C., & Schoenwald, S. K. (2005). The ARC organizational and community intervention strategy for implementing evidence-based children’s mental health treatments. Mental Health Services Research, 7(4), 243–259.

Exchanging Glances?

233

Greenbaum, P. E., Prange, M. E., Friedman, R. M., & Silver, S. E. (1991). Substance abuse prevalence and comorbidity with other psychiatric disorder among adolescents with severe emotional disturbances. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 575–583. Greewald, P., & Cullen, J. W. (1984). The scientiﬁc approach to cancer control. CA – A Cancer Journal for Clinicians, 34, 328–332. Hayes, M. T. (1992). Incrementalism and public policy. New York, NY: Longman. Henggeler, S. W., & Borduin, C. M. (1990). Family therapy and beyond: A multisystemic approach to treating the behavior problems of children and adolescents. Paciﬁc Grove, CA: Brooks/Cole. Henggeler, S. W., Melton, G. B., & Smith, L. A. (1992). Multisystemic treatment of serious juvenile offenders: An effective alternative to incarceration. Journal of Consulting and Clinical Psychology, 60, 953–961. Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., & Cunningham, P. (1998). Multisystemic treatment of antisocial behavior in children and adolescents. In: Treatment manuals for practitioners (p. 287). New York, NY: Guilford Press. Hennggeler, S. W., Schoenwald, S. K., Liao, J. G., Letourneau, E. J., & Edwards, D. L. (2002). Transporting efﬁcacious treatments to ﬁeld settings: The link between supervisory practices and therapist ﬁdelity in MST programs. Journal of Clinical Child and Adolescent Psychology, 31(2), 155–167. Hernandez, M., & Hodges, S. (2001). Theory-based accountability. In: M. Hernandez & S. Hodges (Eds), Developing outcome strategies in children’s mental health. Systems of care for children’s mental health (pp. 21–40). Baltimore, MD: Paul H. Brookes Publishing. Herschell, A. D., McNeil, C. B., & McNeil, D. W. (2004). Clinical child psychology’s progress in disseminating empirically supported treatments. Clinical Psychology: Science and Practice, 11(3), 267–288. Hoagwood, K., Burns, B., Kiser, L., Ringelsen, H., & Schoenwald, S. K. (2001). Evidencebased practice in child and adolescent mental health services. Psychiatric Services, 52, 1179–1189. Hoagwood, K., Burns, B. J., & Weisz, J. R. (2002). A proﬁtable conjunction: From science to service in children’s mental health. In: B. J. Burns & K. Hoagwood (Eds), Community treatment for youth: Evidence-based interventions for severe emotional and behavioral disorders (pp. 327–338). New York, NY: Oxford University Press. Hoagwood, K., Hibbs, E., Brent, D., & Jensen, P. (1995). Efﬁcacy and effectiveness studies of child and adolescent psychotherapy. Journal of Consulting and Clinical Psychology, 63, 683–687. Hoagwood, K., Jensen, P. S., Petti, T., & Burns, B. J. (1996). Outcomes of mental health care for children and adolescents: I. A comprehensive conceptual model. Journal of the American Academy of Child and Adolescent Psychiatry, 35(8), 1055–1063. Hoagwood, K. E. (2005). Family-based services in children’s mental health: A research review and synthesis. Journal of Child Psychology and Psychiatry, 46(7), 690–713. Holden, E. W., De Carolis, G., & Huff, B. (2002). Policy implications of the national evaluation of the comprehensive community mental health services for children and their families program. Children’s Services: Social Policy, Research, and Practice, 5, 57–66. Jensen, P. S. (2003). Commentary: The next generation is overdue. Journal of the American Academy of Child and Adolescent Psychiatry, 42(5), 527–530.

234

ABRAM ROSENBLATT AND LAURA COMPIAN

Jensen, P. S., Weersing, R., Hoagwood, K. E., & Goldman, E. (2005). What is the evidence for evidence-based treatments? A hard look at our soft underbelly. Mental Health Services Research, 7(1), 53–74. Kazdin, A. E. (1999). Current (lack of) status of theory in child and adolescent psychotherapy research. Journal of Child Psychology, 28, 533–543. Kazdin, A. E. (2000). Psychotherapy for children and adolescents: Directions for research and practice. Oxford, UK: Oxford University Press. Kazdin, A. E. (2004). Evidence-based treatments: Challenges and priorities for practice and research. Child and Adolescent Clinics of North America, 13(4), 923–940 [special issue: Evidence-based practice, Part 1: Research update]. Kazdin, A. E. (2006). Arbitrary metrics: implications for identifying evidence-based treatments. American Psychologist, 61(1), 42–49. King, L., Hawe, P., & Wise, M. (1998). Making dissemination a two-way process. Health Promotion International, 13(3), 237–244. Klein, D. F., & Smith, L. B. (1999). Organizational requirements for effective clinical effectiveness studies. Prevention and Treatment, 2, Article 0002a [retrieved November 1, 2005, from journals.apa.org/prevention/volume2/pre0020002a.html]. Knitzer, J. (1982). Unclaimed children. Washington, DC: Children’s Defense Fund. Kutash, K., & Rivera, V. R. (1996). What works in children’s mental health services? Uncovering answers to critical questions. Baltimore, MD: Paul H. Brookes Publishing Company. Lasswell, H. D. (1951). The policy orientation. In: D. Lerneer & H. D. Lasswell (Eds), The policy sciences: Recent developments in scope and methods (pp. 3–15). Stanford, CA: Stanford University Press. Lipkus, I. M., Lyna, P. R., & Rimer, B. K. (1999). Using tailored interventions to enhance smoking cessation among African-Americans at a community health center. Nicotine and Tobacco Research, 1(1), 77–85. Lonigan, C. J., Elbert, J. C., & Johnson, S. B. (1998). Empirically supported psychosocial interventions for children: An overview. Journal of Child Clinical Psychology, 27, 138–145. Manteuffel, B., Stephens, R. L., & Santiago, R. (2002). Overview of the national evaluation of the Comprehensive Community Mental Health Services for Children and their Families Program and summary of current ﬁndings. Children’s Services: Social Policy, Research, and Practice, 5, 3–20. Martell, C. R., & Hollon, S. D. (2001). Working together on shifting ground: Researcher and clinician collaboration in clinical trials. Behavior Therapist, 24(7), 144–146. Martin, G., Herie, M., Turner, B., & Cunningham, J. (1998). A social marketing model for disseminating research-based treatments to addictions treatment providers. Addiction, 93, 1703–1715. Milio, N. (1983). Primary care and the public’s mental health. Lexington, MA: Lexington Books. Nathan, P. E. (2004). The evidence base for evidence-based mental health treatments: Four continuing controversies. Brief Treatment and Crisis Intervention, 4(3), 243–254. National Advisory Mental Health Council. (1999). Bridging science and service: A report by the National Advisory Mental Health Council’s Clinical Treatment and Services Research Workgroup. Bethesda, MD: National Institute of Mental Health. National Institute of Mental Health. (1983). Program announcement: Child and adolescent service system program. Rockville, Maryland: Author.

Exchanging Glances?

235

Newman, M. G., & Castonguay, L. G. (1999). Reﬂecting on current challenges and future directions in psychotherapy: What can be learned from dialogues between clinicians, researchers, and policy makers? Journal of Clinical Psychology, 55, 1407–1413. Ollendick, T. H., & Davis, T. E., III. (2004). Empirically supported treatments for children and adolescents: Where to from here? Clinical Psychology: Science and Practice, 11(3), 289–294. Rakowski, W., Lipkus, I. M., Clark, M. A., Rimer, B. K., Ehrich, B., Lyna, P. R., & Kornguth, P. J. (2003). Reminder letter, tailored step-care, and self-choice comparison for repeat mammography. American Journal of Preventive Medicine, 25(4), 308–314. Rosenblatt, A. (1993). In home, in school, and out of trouble. Journal of Child and Family Studies, 2(4), 275–282. Rosenblatt, A. (1998). Assessing the child and family outcomes of systems of care for youth with severe emotional disturbance. In: M. H. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: Programs and evaluation best practices (pp. 329–362). Austin, TX: PRO-ED. Rosenblatt, A. (2005). Assessing child and family outcomes. In: M. H. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: Programs and evaluation best practices. Austin, TX: PRO-ED (2nd ed.). Rosenblatt, A., & Attkisson, C. C. (1997). Integrating systems of care in California for youth with severe emotional disturbance. IV: Educational attendance and achievement. Journal of Child and Family Studies, 6(1), 113–129. Rosenblatt, A., & Rosenblatt, J. (2000). Demographic, clinical and functional characteristics of youth enrolled in six California systems of care. Journal of Child and Family Studies, 9(1), 51–66. Rosenblatt, A., & Woodbridge, M. W. (2003). Deconstructing research on systems of care for youth with emotional and behavioral disorders: Frameworks for policy research. Journal of Emotional and Behavioral Disorders, 11(1), 27–37. Rosenblatt, J. A., Rosenblatt, A., & Biggs, E. E. (2000). Criminal behavior and emotional disorder: Comparing youth served by the mental health and juvenile justice systems. Journal of Behavior Health Services and Research, 27(2), 227–237. Schoenwald, S. K., Borduin, C. M., & Henggeler, S. W. (1998). Multisystemic therapy: Changing the natural and service ecologies of adolescents and families. In: M. Epstein, K. Kutash & A. Duchnowski (Eds), Outcomes for children and youth with emotional and behavioral disorders and their families: Programs and evaluation best practices (pp. 485–511). Austin, TX: PRO-ED. Schoenwald, S. K., & Hennggeler, S. W. (2004). A public health perspective on the transport of evidence-based practices. Clinical Psychology: Science and Practice, 11(4), 360–363. Schoenwald, S. K., Sheidow, A. J., & Letourneau, E. J. (2004). Toward effective quality assurance in evidence-based practice: Links between expert consultation, therapist ﬁdelity, and child outcomes. Journal of Clinical Child and Adolescent Psychology, 33(1), 94–104. Schoenwald, S. K., Sheidow, A. J., Letourneau, E. J., & Liao, J. G. (2003). Transportability of multisystemic therapy: Evidence for multilevel inﬂuences. Mental Health Services Research, 5(4), 223–239. Scriven, M. (1967). The methodology of evaluation. In: Perspectives of curriculum evaluation. AERA monograph. Chicago, IL: Rand McNally.

236

ABRAM ROSENBLATT AND LAURA COMPIAN

Sechrest, L., & Rosenblatt, A. (1987). Research methods. In: H. Quay (Ed.), Handbook of juvenile delinquency. Wiley series on personality processes (pp. 417–450). Oxford, UK: Wiley. Sechrest, L., West, S. G., Phillips, M. A., Redner, R., & Yeaton, W. (1979). Some neglected problems in evaluation research: Strength and integrity of treatments. In: L. Sechrest, S. West, M. Phillips, R. Redner & W. Yeaton (Eds), Evaluation studies review annual, (Vol. 4, pp. 15–35). Beverly Hills, CA: Sage Publications. Silverman, W. K., Kurtines, M., & Hoagwood, K. (2004). Research progress and effectiveness, transportability, and dissemination of empirically supported treatments: Integrating theory into research. Clinical Psychology: Science and Practice, 11, 295–299. Sorensen, J. L., Hall, S. M., Loeb, P., Allen, T., Glaser, E. M., & Greenberg, P. D. (1988). Dissemination of a job seekers’ workshop to drug treatment programs. Behavior Therapy, 19, 143–155. Southam-Gerow, M. A., Weisz, J. R., & Kendall, P. C. (2003). Youth and anxiety disorders in research and service clinics: Examining client differences and similarities. Journal of Clinical Child and Adolescent Psychology, 32, 375–385. Stirman, S. W., Crits-Christoph, P., & DeRubeis, R. J. (2004). Achieving successful dissemination of empirically supported psychotherapies: A synthesis of dissemination theory. Clinical Psychology: Science and Practice, 11(4), 343–359. Stroul, B. A. (1993). Systems of care for children and adolescents with severe emotional disturbances: What are the results? Washington, DC: Georgetown University Child Development Center. Stroul, B. A., & Friedman, R. M. (1986). A system of care for seriously emotionally disturbed children and youth. Washington, DC: Georgetown University Child Development Center. Stroul, B. A., & Friedman, R. M. (1996). The system of care concept and philosophy. In: B. A. Stroul (Ed.), Children’s mental health: Creating systems of care in a changing society. Baltimore, MD: Paul H. Brookes Publishing. Stroul, B. A., Pires, S. A., Armstrong, M. I., & Zaro, S. (2002). The impact of managed care on system of care that serve children with serious emotional disturbances and their families. Children’s Services: Social Policy, Research, and Practice, 5(1), 21–36. Task Force on Promotion and Dissemination of Psychological Procedures, Division of Clinical Psychologist, American Psychological Association. (1995). Training in and dissemination of empirically-validated psychological treatments: Report and recommendations. The Clinical Psychologist, 48, 3–23. Task Force on Psychological Intervention Guidelines, American Psychological Association. (1995). Template for developing guidelines: Interventions for mental disorders and psychosocial aspects of physical disorders. Washington, DC: American Psychological Association. Tharp, R. G., & Gallimore, R. (1982). Inquiry process in program development. Journal of Community Psychology, 10(2), 103–118. Tolan, P. H., & Dodge, K. A. (2005). Children’s mental health as a primary care and concern: A system for comprehensive support and service. American Psychologist, 60(6), 601–614. U.S. Department of Health and Human Services. (1999). Mental health: A report of the surgeon general. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health.

Exchanging Glances?

237

VanDenBerg, J. E., & Grealish, E. M. (1996). Individualized services and supports through the wraparound process: Philosophy and procedures. Journal of Children Family Studies, 5(1), 7–21. Weisz, J. R. (2000). Agenda for child and adolescent psychotherapy research: On the need to put science into practice. Archives of General Psychiatry, 57, 837–838. Weisz, J. R., Chu, B. C., & Polo, A. J. (2004). Treatment dissemination and evidence-based practice: Strengthening intervention through clinician–researcher Collaboration. Clinical Psychology: Science and Practice, 11, 300–3007. Weisz, J. R., Donenberg, G. R., Han, S. S., & Weiss, B. (1995a). Bridging the gap between laboratory and clinic in child and adolescent psychotherapy. Journal of Consulting and Clinical Psychology, 63, 688–701. Weisz, J. R., Huey, S. J., & Weersing, V. R. (1998). Psychotherapy outcome research with children and adolescents: The state of the art. Advances in Clinical Child Psychology, 20, 49–91. Weisz, J. R., & Jensen, A. L. (1999). Efﬁcacy and effectiveness of child and adolescent psychotherapy and pharmacotherapy. Mental Health Services Research, 1, 125–157. Weisz, J. R., & Jensen, A. L. (2001). Child and adolescent psychotherapy in research and practice contexts: Review of the evidence and suggestions for improving the ﬁeld. European Child and Adolescent Psychiatry, 10, 12–18. Weisz, J. R., Jensen, A. L., & McLeod, B. D. (2005). Development and dissemination of child and adolescent psychotherapies: Milestones, methods, and a new deployment-focused model. In: E. Hibbs & P. Jensen (Eds), Psychosocial treatments for child and adolescent disorders: Empirically based strategies for clinical practice. Washington, DC: American Psychological Association. Weisz, J. R., Sandler, I. N., Durlak, J. A., & Anton, B. S. (2005). Promoting and protecting youth mental health through evidence-based prevention and treatment. American Psychologist, 60(6), 628–648. Weisz, J. R., Weiss, B., Alicke, M. D., & Klotz, M. L. (1987). Effectiveness of psychotherapy with children and adolescents: A meta-analysis for clinicians. Journal of Consulting and Clinical Psychology, 55, 542–549. Weisz, J. R., Weiss, B., Han, S. S., Granger, D. A., & Morton, T. (1995b). Effects of psychotherapy with children and adolescents revisited: A meta-analysis for clinicians. Journal of Consulting and Clinical Psychology, 55, 542–549. Wennberg, J. E. (1990). Small area analysis and the medical care outcome problem. In: L. Sechrest, E. Perrin & J. Bunker (Eds), Research methodology: Strengthening causal interpretations of nonexperimental data. Rockville, MD: Agency for Health Care Policy and Research.

Mental Health Services For Minority Ethnic Children And Adolescents (Child and Adolescent Mental Health)

The Economics of Obesity, Volume 17 (Advances in Health Economics and Health Services Research) (Advances in Health Economics and Health Services Research) ... Economics and Health Services Research)

Research on Community-Based Mental Health Services for Children and Adolescents, Volume 14 (Research in Community and Mental Health)

Mental Health Services For Minority Ethnic Children And Adolescents (Child and Adolescent Mental Health)

Children and adolescents with mental health problems

Mental Health Treatment for Children and Adolescents (Evidence-Based Practice)

Methods for Disaster Mental Health Research

Neuroimaging Research in Geriatric Mental Health

Choosing Methods in Mental Health Research: Mental Health Research from Theory to Practice

Mental Health Services: A Public Health Perspective

Religion and Mental Health

Spirituality and Mental Health

Mental Health and Disasters

Disasters and Mental Health

Confidentiality and mental health

Mental Health and Inequality

Migration and Mental Health

Disasters and Mental Health

The Economics of Obesity, Volume 17 (Advances in Health Economics and Health Services Research) (Advances in Health Economics and Health Services Research) ... Economics and Health Services Research)

Emergency department handbook: children and adolescents with mental health problems

Quality of Life and Mental Health Services

Planning and Budgeting to Deliver Services for Mental Health (Mental Health Policy and Service Guidance Package)

International Perspectives on Children and Mental Health 2 volumes (Child Psychology and Mental Health)

Handbook of Mental Health Services for Children, Adolescents, and Families (Issues in Clinical Child Psychology)

Handbook of Mental Health Services for Children, Adolescents, and Families (Issues in Clinical Child Psychology)

Understanding Children and Young People's Mental Health

International Perspectives on Children and Mental Health 2 volumes (Child Psychology and Mental Health)

Developing Mental Health Services for Children and Adolescents with Learning Disabilities: A Toolkit for Clinicians

Teamworking in Mental Health

Mental Health Financing (Mental Health Policy and Service Guidance Package)

Mental Health Context (Mental Health Policy and Service Guidance Package)

Clinical Management in Mental Health Services

Clinical Management in Mental Health Services

Research on Community-Based Mental Health Services for Children and Adolescents, Volume 14 (Research in Community and Mental Health)

Mental Health Services For Minority Ethnic Children And Adolescents (Child and Adolescent Mental Health)

Children and adolescents with mental health problems

Mental Health Treatment for Children and Adolescents (Evidence-Based Practice)

Methods for Disaster Mental Health Research

Neuroimaging Research in Geriatric Mental Health

Choosing Methods in Mental Health Research: Mental Health Research from Theory to Practice

Mental Health Services: A Public Health Perspective

Religion and Mental Health

Spirituality and Mental Health

Mental Health and Disasters

Disasters and Mental Health

Confidentiality and mental health

Mental Health and Inequality

Migration and Mental Health

Disasters and Mental Health

The Economics of Obesity, Volume 17 (Advances in Health Economics and Health Services Research) (Advances in Health Economics and Health Services Research) ... Economics and Health Services Research)

Emergency department handbook: children and adolescents with mental health problems

Quality of Life and Mental Health Services

Planning and Budgeting to Deliver Services for Mental Health (Mental Health Policy and Service Guidance Package)

International Perspectives on Children and Mental Health 2 volumes (Child Psychology and Mental Health)

Handbook of Mental Health Services for Children, Adolescents, and Families (Issues in Clinical Child Psychology)

Handbook of Mental Health Services for Children, Adolescents, and Families (Issues in Clinical Child Psychology)

Understanding Children and Young People's Mental Health

International Perspectives on Children and Mental Health 2 volumes (Child Psychology and Mental Health)

Developing Mental Health Services for Children and Adolescents with Learning Disabilities: A Toolkit for Clinicians

Teamworking in Mental Health

Mental Health Financing (Mental Health Policy and Service Guidance Package)

Mental Health Context (Mental Health Policy and Service Guidance Package)

Clinical Management in Mental Health Services

Clinical Management in Mental Health Services

Recommend Documents