Weighing evidence for common and specific factors in psychotherapy with children

Since publication of the first edition of Persuasion and Healing (1961), our understanding of the necessary ingredients of effective psychotherapy has grown substantially. In the main, the book's key insights about the centrality of empathic, healing relationships, informed by keen appreciation and respect for patients' cultural contexts, have withstood rigorous scrutiny. Arguably, these concepts apply even more strongly to child psychotherapies, in which effectiveness depends both on the therapist's relationships with the child and the parents and on the therapist's understanding of the cultural context of the child's family. Though child and adolescent psychotherapy research has lagged behind studies in adults, the literature pertaining to the treatment of children's mental disorders has grown exponentially in recent years (Wang et al. 2003). This surge of publications has been a mixed blessing. In many ways, the field resembles the entrepreneurial decades of the 1960s and 1970s, when adult therapies ranging from abreaction to Zen far outran careful research meant to develop or evaluate them. More than 90 percent of the more than 550 "psychotherapies" claimed as useful to treat childhood disorders have not been studied empirically (Kazdin 2000). The field of child psychotherapy cries out for rigorous studies of the efficacy and effectiveness of both common and novel treatments. About ten years before Persuasion and Healing, Eysenck (1952) asserted that psychotherapeutic practices were no more effective than the passage of time, a claim that Frank and others repeatedly refuted, citing evidence from numerous well-designed studies in adults. Nevertheless, early reviews of the child psychotherapy literature (Levitt 1957, 1963) supported Eysenck's assertion and led many clinicians and researchers to question the efficacy of psychotherapy for both adults and children. Even in more recent years, Bickman and colleagues noted that a large, expensive attempt to provide optimal mental health services for children and adolescents was of no more benefit than usual care provided to a control group, combined with the passage of time-the so-called clocksetting cure (Bickman et al. 1995; Bickman, Noser, and Summerfelt 1999; Lambert and Bickman 2004). Notably, the early discouraging reviews stimulated valuable further studies of developmental psychopathology, diagnostic nomenclature, and assessment and treatment of children and adolescents. Researchers now typically focus on the ingredients of effective care, not just how it is organized or whether it is empathic and "family friendly." Current research in child psychotherapy seeks to determine the conditions under which certain treatments are effective, moving past the older question of whether therapy is effective. This has led to an exponential increase in studies of treatment process and outcome, using more rigorous experimental designs (Durlak et al. 1995; Kazdin 2000). Four major meta-analyses examining the effects of child psychotherapy (Casey and Berman 1985; Weisz, Weiss, Alicke, and Klotz 1987; Weisz, Weiss, Han, et al. 1995) counter the findings of Lambert and Bickman, offering good evidence that certain forms of child psychotherapy are beneficial for children and their families. Specifically, the current research suggests that some of the psychotherapies for children may be superior to waiting-list and attention-placebo conditions, although exceptions abound (Jensen et al. 2005). Merely demonstrating the efficacy of therapy in general or of one particular method against placebo leaves many questions unanswered. Persuasion and Healing was prescient in highlighting the diffi cult next step: diff erentiating among therapies and demonstrating that specifi c approaches might be best for particular conditions. Every edition of the book evaluated new claims of those who touted the superiority of their brand of treatment for adults. In the end, the book adopted the agnostic position that, with the exception of exposure for anxiety, such assertions were unproven. Meta-analyses published in the decades since the third edition of Persuasion and Healing (Frank and Frank 1991) provide new evidence that some forms of psychotherapy work to correct certain behavioral, emotional, and social problems in children of particular ages, under specified conditions. This step necessarily precedes that of demonstrating that any particular therapy outperforms any other active therapy. "Comparative effectiveness" research, much sought by policymakers, remains an unattainable goal for the field, with the possible exception of the few studies that compare psychotherapy with medication. This chapter briefly reviews the current state of research in child and adolescent psychotherapy (see also chapters 1 and 11). Specifi cally, we attempt to document the movement toward evidence-based practices, defined here as evidence derived from randomized controlled trials (RCTs) (Ollendick and King 2004). In RCTs, patient-subjects are randomly assigned to treatment or to some control condition, such as a waiting-list, attention-placebo condition or an alternative form of treatment. After defining empirically supported treatments and examining their current status, we focus on certain trends emerging from a systematic search of the literature. In particular, we emphasize recent investigations that compile reports on types and targets of treatment. Next, we illustrate and discuss current issues associated with evidence-based treatments and their development. Beyond discussing differential effects of various treatments, we examine the use of treatment manuals, relationship variables, and the challenge of aligning clinical practice with rigorously tested, evidence-based approaches.