'Pilot RCT of a new treatment for child conduct problems that have not responded to evidence-based parent training'.

Abstract

Children with conduct problems are at high risk of persisting mental health problems, making them a priority for early intervention. Group-based parent-training is effective but with a substantial failure rate. Based on evidence on the value of involving children, we developed Reflective Interpersonal Therapy for Children and Parents, (RICAP). We report feasibility and outcomes from the first pilot trial of an intervention for children with conduct problems persisting after parent training. In contrast to most studies, we used both parent and teacher report. The sample comprised 105 children and their parents aged 4-10 years inclusive referred to UK Child and Adolescent Mental Health Services (CAMHS) with conduct problems. All were offered the Incredible Years (IY) parent training intervention, and parents provided pre- and post-treatment measures (including CBCL, SDQ). Children still above clinical threshold after IY were randomized either to RICAP or to usual CAMHS treatment (CTAU) with follow up 8 months later. Only those randomized to CTAU could receive school-based interventions. Trial Registration Number: ISRCTN25252940. Feasibility was supported by high retention through the initial IY (102/105) and subsequent RCT phases of the study (58/70 eligible for randomization). The majority of those randomized to RICAP attended for 11/14 or more sessions, reflecting its high acceptability to both children and parents. By parent report RICAP was superior to CTAU on CBCL externalising (d = 0.32) and internalising (d = 0.42) problems, while by teacher report CTAU was superior on SDQ total problems (d = 0.32) and reactive aggression (d = 0.27). We provide first evidence of the feasibility of a design in which non-responders to parent training are randomized to novel intervention or control conditions. Outcomes were different for parent and teacher reports, highlighting for future randomized controlled trials the need to attend to possible reporter and treatment context variations.