III
Targeted Problem Intervention Programs
Chapter 7
Characteristics Shared by Exemplary Child Clinical Interventions for Indicated Populations
Patrick H. Tolan
University of Illinois at Chicago
This section of this volume describes direct service interventions for children with current emotional and behavioral pathology. All of these exemplary intervention programs share an intention of directly affecting the current and future symptomatology of indicated populations (Gordon, 1983; Tolan & Guerra, 1994a). Direct interventions are the focus of the majority of service resources, funding, and child clinician efforts in our society, at this time (Kazdin, 1987; Tuma, 1989). Thus, a very important question is, what characteristics are common to exemplary direct service programs? Identifying such characteristics can help us understand essential qualities for apt service delivery, and can increase the efficiency of program development.
As one reads the description of these exemplary programs and consults other pertinent literature an impressive and somewhat surprising pattern of characteristics emerges. The program design and development characteristics are remarkably consistent across the programs described in this section of the book, despite their varying approaches and different interests. These characteristics also differentiate these programs from more typical service delivery systems. It may be that these characteristics are the framework for efficacious direct intervention for children and may be necessary for effective service.1 If so, then these characteristics may be requisite considerations for design of service systems, research, and evaluation. This chapter attempts to articulate what these characteristics are, note and specify the examples of such characteristics among the represented intervention programs, recognize impediments to implementation of them to general service models, and suggest needed further developments.
UNDERSERVICE AND MISAPPLIED SERVICES
There is clear and urgent need for better understanding of what are effective and useful clinical services. The consensus is that current service delivery systems are inadequate (Knitzer, 1982). The current practices underserve children and families in need in several ways. Many children in need of clinical intervention are not receiving services. For example, Weisz, Weiss, and Donenberg (1992) estimated that at any time 12% of the 63 million children in the United States suffer from serious behavioral or emotional problems (citing Institute of Medicine, 1989; Saxe, Cross, & Silverman, 1988). However, only about 2.5% receive treatment (Office of Technology Assessment, 1986). Knitzer (1982) found that 20% of children with DSM-III psychiatric symptoms were receiving treatment and Tuma (1989) estimated that only 4% to 6% of children in the U.S. have at least one outpatient visit each year even though 20% to 30% are in need of such aid. It is not just that children in need are not getting serviceāsome services are misapplied. For example, Offer, Ostrov, and Howard (1987) found that almost one third of adolescents receiving psychotherapy did not evidence substantial clinical symptoms when administered an independent measure of psychopathology.
In addition to undersupply and misdirection of service, service is unevenly accessible. This differential availability depends on the severity and type of symptoms, the social and economic resources of the family, and the nature of the contact referral (John, Offord, Boyle, & Racine, 1995; Kazdin, 1989). For example, children with conduct disorders and other behavioral problems are disproportionately referred for mental health services, compared to those with other disorders (John et al., 1995; Weiss et al., 1973). Also, children referred by affiliated services or with co-occurring medical conditions are more likely to use services (John et al., 1995; Tolan, Ryan, & Jaffe, 1988). Children coercively referred (the majority), are less likely to continue (Viale-Val, Rosenthal, Curtiss, & Marohn, 1984). In regard to social and economic influences, poorer children are more likely to be referred for psychiatric services than are higher socioeconomic status (SES) children, given equal availability of services (John et al., 1995; Weiss et al., 1973). Children from low-income families are overrepresented among clinicsā clients (Garralda & Bailey, 1988; Weiss et al., 1973). However, when access is not equalized, higher maternal education relates to greater use of mental health services for children (Mitchell & Smith, 1981). These influences on differential access create a service pattern of those children with more severe problems often being the least served, those with the least other resources having the least access to mental health services, and those most in need least likely to access and retain adequate care (Tolan et al., 1988).
Even if access barriers can be overcome there are service system characteristics that diminish service adequacy (Knitzer, 1982). Fortunately, programs such as those included in this volume represent preferable alternatives. The isolation of child mental health services and psychological developmental issues from the purview of general pediatric care is one example (see Olson & Netherton; Schroeder, this volume). Other examples are the poor coordination of health care, education, and social service systems (see Hannah & Nichol and Klingman, this volume), the poor integration of services for severe and chronic problems (see Pelham et al.; Kirigin; Mesibov; this volume) and the emphasis on aftercare and crisis intervention over early intervention and preventive orientation (see Lavigne, this volume). As each of the exemplary chapters in this section illustrates, these structural barriers to service delivery can be overcome.
Even when contact is made to obtain psychotherapeutic services, attendance rates beyond contact are usually about 50% to 70% (Tuma, 1989). In one study of an adolescent outpatient clinic, Tolan et al. (1988) found that 48% of families contacting a clinic for psychotherapy actually kept their first appointment, and 58% of those with follow-up contact continued beyond diagnostic sessions. In addition, as others have found, in that study continuation was related to type of referral problem, the continuity of caregiver, and inclusion of the family (John et al., 1995; Tolan et al., 1988). It appears that, even in innovative service systems, the more severe cases are not as likely to receive full and extended services (Pillen & Tolan, 1994).
Another aspect of underservice to children is that services provided are not determined by the problems presented or influenced by the circumstances of the family (Kruesi & Tolan, in press). Clients are apt to receive services that are based on a clinic philosophy rather than the specific problem they bring. For example, if they happen to contact a behavioral clinic or practitioner they will receive that approach; if they contact a family therapist they will be offered that approach (Kazdin, 1987; Kazdin, Siegel, & Bass, 1990; Tuma, 1989). This creates a situation of clients getting service based on what the clinicians happen to know or philosophically prefer. Even though eclecticism is endorsed as the perspective of the majority of practicing child clinicians, the rationale for and specificity of use of different methods is less clear (Hiatt & Goldman, 1994). Also, although most child clinicians endorse research-based practice, few rely on research and many describe it as irrelevant (Kazdin et al., 1993). The result is that two children with the same problem are likely to receive very different services depending on the clinic they go to, and two children with quite dissimilar problems may receive quite similar interventions simply because they came to the same clinic.
When methods used and the organization of services are not problem-oriented, they are less likely to be developmentally attuned and offered in modes or on schedules that fit the clientās needs (Kazdin, 1989; Tuma, 1989). Instead clinic procedures and case formulations follow standardized practices due to traditions and ease of scheduling. This problem in childrenās mental health services has been aptly termed chronic undifferentiated treatment (Leventhal, personal communication, 1994). Treatment needs to be more specific to the problem children and families present, have a clear articulation of the theoretical model of psychopathology, therapistāclient relationship, and how change occurs (Kazdin, Siegel, & Bass, 1990; Tolan, 1991) and be responsive to the circumstances of clientsā lives (Day & Roberts, 1991; Knitzer, Steinberg, & Fleisch, 1990; Kruesi & Tolan, in press; Roberts & Hinton-Nelson, this volume).
The overall effect of these and other inadequacies is that many children with emotional and behavioral problems in need of clinical intervention are not receiving such aid and those coming to many clinics and practitioners are not receiving apt treatment (Cohen, 1995; Knitzer et al., 1990). In part, there is need for reform of larger system structures, including financing of service provision and criteria for access to mental health services (Knitzer et al., 1990). In addition, there is need for reevaluation of how service is provided, what service is provided for what problems (how diagnostic and triage decisions are made), and how to make mental health service systems more user friendly. These smaller system reforms are also easier to institute than is the massive shift in the economic base and health system planning required to reorient overall service to children (Cohen, 1995; Hiatt & Goldman, 1994). The interventions collectively described here suggest some directions for developing effective service methods, delivering them efficiently, and doing so in a manner that is accessible to those in need.
EFFICACY EVIDENCE AND COMPONENT ORGANIZATION
In contrast to these critical shortcomings of service delivery and prevailing practices, there is substantial evidence that direct interventions meant to have psychotherapeutic impact are beneficial to the majority of clients if carefully and properly delivered (Casey & Berman, 1985; Kazdin, Bass, Ayers, & Rodgers, 1990; Weisz, Weiss, & Donenberg, 1992; Weisz, Weiss, Alicke, & Klotz, 1987). In meta-analyses, treatment effect of psychotherapy services ranged from M = .71 (Casey & Berman, 1985) to M = .89 (Weisz et al., 1987). This range of effect sizes mean that on average between 76% and 90% of child clients receiving psychotherapy have better post-treatment status than untreated comparisons. These effect sizes are on a par with those found for adults and as large if not larger than many common medical procedures (Lipsey & Wilson, 1993). Such substantial levels of effect are drawn from a relatively equal number of studies of externalizing and internalizing problems, and age groups, genders, and therapist training levels and are found across types of problems (Kazdin, Seigel, & Bass, 1992; Weisz et al., 1987).
However, the efficacy demonstrations that are the basis of the metaanalysis effect sizes differ from common clinical practices in several ways. They were more likely to be (a) behavioral or cognitiveābehavioral than other approaches; (b) to use specially trained therapists, (c) have carefully defined inclusion rules, (d) recruit participants, and (e) provide manualized or protocol-based interventions than occurs in practice (Weiss & Weisz, 1990; Weisz et al., 1995). Thus, as noted by Weisz, Weiss, Morton, Granger, and Han (1992), their generalization to common practice may be limited. Also, as noted by Kazdin et al. (1992), the available studies do not include evaluations of most of the wide variety of psychotherapeutic methods being employed with children. The net effect is the database is built from tests of few approaches as applied to a few problems, with only a few instances of more than one evaluation of a given approach to a given problem. Our field has demonstrated efficacy of some treatment(s) for most childhood and adolescent problems, but lacks a clear understanding of their preferability over other approaches and how that preferability varies by problem and family circumstance (Goldman et al., 1990). Also, the generalization to general practice is unclear.
In other words, even though emotional state, behavior, and attitude change can be demonstrated under controlled field experiment conditions, tests of clinical services as normally delivered do not show any empirically demonstrated benefits of psychotherapeutic interventions (Weisz et al., 1995). The common conclusion drawn from the contrast between the efficacy demonstrations and failure to find effects in clinical evaluations is that the demonstrationās empirical findings do not translate to the āreal world.ā However, it may be that the current methods of service delivery impede effective service and that the difference is due to organization and specificity of services (Cohler & Tolan, 1993; Hiatt & Goldman, 1994). Therapist training including previous experience may often be too general to permit the development of expertise with specific problems that is essential to regular success.
Eclectic thinking may inhibit selection of methods by presenting problem, developmental considerations, or family circumstance, and therefore may reduce efficacy (Tolan, 1991). The gap between demonstrated controlled trials and the failure to evidence effectiveness of services as usually delivered may not be found due to artifacts of design. It may be that these unrealistic characteristics are indications of service delivery structures that prevent efficacious approaches being adequately delivered. Evidence that this is the case can be found in the types of clinical approaches and programs presented in the ensuing chapters in this section. Thus the needed change to realize effective services is to develop service delivery based on the principles described here including particular attention to an empirical base using a problem-oriented approach to organizing services.
The programs described in this section vary in theoretical orientation, problem of interest, and orientation to research. However, they share characteristics that can guide us towards apt and effective direct mental health services to children. They represent demonstrations of psychotherapeutic work that is problem-oriented and use a set of practices based on a theoretical and empirical base. Each chapter describes program development based on a critical evaluation approach to components and service delivery model. Although this set of reports includes interventions that have not been fully validated, each is based in scientific literature and empirical findings. Also, each has a theoretical rationale that is directly drawn from the extant literature on behavior change, intervention design, service delivery, and developmental psychopathology. They represent evidence that something can be done about most of the troublesome problems that children bring to mental health service agencies (Schorr & Schorr, 1988). The development of each described program has been a process of eliminating some practices and a justification of retained efforts because of theoretical and empirical evidence. Their shared characteristics seem to comprise a guide for more effective mental health services and more accessible and apt service delivery to children.
SHARED CHARACTERISTICS OF EXEMPLARY INDICATED PROGRAMS
Many of the shared characteristics of importance were identified by Roberts and Hinton-Nelson in the introductory chapter of this volume. They are principles that mark quality across types of intervention. Thus, each program in this section of the book is guided by clearly defined missions and philosophies. However, th...