Treatment of Severe Childhood Aggression (TOSCA) Studies

For decades we have known that physical aggression in youth often has dire consequences in later life. The Treatment of Severe Childhood Aggression or TOSCA study was a departure from earlier monotherapy studies of medication treatment for aggression because it entailed an attempt to “normalize” behavioral outcomes through the use of combined treatment, if needed. The impetus for TOSCA was a series of risperidone studies conducted by Janssen Pharmaceutica and others (e.g., Findling et al., 2000) published in the early 2000s. When designing TOSCA, we were especially interested in risperidone's potential to dampen aggression and other disruptive behaviors. Initially, our focus was comparison of psychostimulant vs. risperidone monotherapy, both with an adjunctive parent training (PT) program. However, using the best data available, power analyses indicated that over 2,000 participants would be needed to have sufficient sensitivity to separate treatments! Eventually, we settled on a combined treatment design not only because it dealt with the statistical-power issue but also because it was ecologically valid (i.e., it followed a common clinical practice; see Farmer et al. 2011 for methodological details).

TOSCA is important because it has mapped uncharted territory insofar as pediatric drug research design is concerned. To satisfy the statistical-power issues, we advocated for the complex add-on design (i.e., combined therapy) despite the fact that we, like most practitioners, were trained that “simpler is better.” The TOSCA design not only called for the use of combined treatments, but one of those treatments was an antipsychotic. Such agents often provoke concern because of their reputation as “powerful drugs,” because they can cause possible long-term movement disorders, and because of their association with metabolic syndrome. Eventually, the NIMH funded TOSCA and, in so doing, acknowledged the importance of aggression in today's society and the reality that combined pharmacotherapy has become a common practice in child and adolescent psychiatry.

The TOSCA participants were 168 children with severe physical aggression (to others, to self, and/or to property) and presence of attention-deficit/hyperactivity disorder (ADHD); psychotic and bipolar disorders were excluded. All study participants received psychostimulant medication and PT in behavior management. If disruptive behavior was not normalized by 3 weeks, then either placebo or risperidone was added as co-therapy from weeks 4 through 9. Psychostimulant+PT+placebo were termed basic treatment, whereas psychostimulant+PT+risperidone were termed augmented treatment. Our earliest reports indicated that augmented treatment was more effective than basic treatment in reducing parent ratings of disruptive behavior (moderate effect size) but not clinician global ratings.

Three TOSCA studies appear in this issue. Arnold et al. (2015) examined domains other than disruptive behavior as rated by parents and teachers. They found that augmented treatment resulted in a reduction to teacher-rated anxiety and further that these reductions mediated improvements in parent-rated disruptive behavior. Farmer et al. (2015) looked for predictors and moderators of treatment outcome. They found that “callous/unemotional” traits and ADHD symptom severity were predictors of outcome, whereas anger/irritability symptoms, manic scores, and (importantly) maternal education variously moderated changes in disruptive or prosocial behavior. Equally important was the failure of our proactive and reactive aggression measure to moderate outcomes. Finally, Rundberg-Rivera et al. (2015) surveyed 150 of the participating families and reported remarkably high levels of satisfaction with the research experience (e.g., 98% said that they would recommend the study to other parents of children having similar difficulties). Parents enthusiastically endorsed the PT provided, and it seemed to be an important part of the TOSCA experience. We strongly recommend PT or similar psychosocial support in future trials even though such a successful control condition may well make it difficult to show further benefit from any additional treatments being assessed.

TOSCA has provided several answers to important questions regarding treatment of serious physical aggression in children. In forthcoming reports, we shall describe maintenance of therapeutic gains, treatment effects on learning performance, and observable changes in a social interaction paradigm between child and parent.