Longitudinal studies have shown that adolescent and adult antisocial behavior and criminal involvement often have roots in conduct problems that begin in early childhood. Both individual risk factors (e.g., undercontrolled temperament, attention problems and delayed motor development) and social factors (e.g. parental neglect of child, inconsistent and harsh discipline) are important to the emergence and persistence of conduct problems [
1]. Fortunately, extensive research has shown that evidence-based parent training interventions are effective in reducing child conduct problems [
2],[
3]. However, the effects of parent training do not necessarily diffuse to settings other than the home [
4],[
5]. Interventions that directly target children are important when parents can or will not participate in parent training due to contextual factors, such as life stress, work conflicts, family issues, interpersonal issues and parental psychopathology [
6],[
7]. Furthermore, children with conduct problems are at risk of being rejected and disliked by peer students as early as preschool and have been found to actively seek peers who are behaviorally similar to themselves [
8]. These findings have inspired the development of Individual Social Skills Training (ISST), which is one of the interventions in the multi-modular program Early Initiatives for Children at Risk (Norwegian acronym, TIBIR). This program was developed with the aim of scaling up the use of evidence-based interventions for the prevention and reduction of conduct problems in children [
9]. In addition to ISST, TIBIR also consists of three parent training interventions and one teacher training intervention [
9]. These interventions are tailored based on the severity of the children’s conduct problems. ISST is always offered in combination with parent training, as TIBIR builds on the principles of the social interaction learning (SIL) model, suggesting that conduct problems are caused by coercive and aggressive parent-child interactions [
10]. Another overriding principle of TIBIR is that all interventions should be evidence-based; currently three have been evaluated in effectiveness trials [
4],[
11],[
12]. Therefore, in an effort to evaluate the effects of every intervention in TIBIR, the present study tested the unique effectiveness of ISST (without parent or teacher training) in a randomized, controlled trial immediately following and six months after the completion of the intervention. The fact that ISST was tested in an effectiveness trial (i.e., real world settings), and not in an efficacy trial (i.e., optimal settings), makes it likely that the findings from the current sample are generalizable to the population of Norwegian children with conduct problems.
Individual social skills training
Social skills training is often conducted as a group intervention in which children are taught social and cognitive skills (i.e., play, friendship and conversational skills, problem solving, self-control, anger management, empathy training, and perspective taking). However, social skills training may also be individually delivered to children to avoid negative group influences. Group-based social skills training in homogeneous groups of children and youth with conduct problems may produce unintended negative outcomes; this result is often referred to as deviance training [
13],[
14].
Individual training by
coaching is designed to teach children the skills they need for social acceptance and friendship through techniques such as discussion, rehearsal, and feedback from the coach. The advantage of individual training is that skills may be tailored and selected after considering what skills would be most important for the young person to learn. Based on the theories underlying social skills training, it has been assumed that learning individual social skills should result in a reduction of externalizing problem behavior. However, the empirical support for this assumption has turned out to be rather weak [
15],[
16].
Research on social skills training
In general, the findings regarding child social skills training interventions have been mixed. Some have been positive, such as the findings that showed that children who received The Incredible Years child training had significantly lower parent-reported externalizing problems and less teacher-reported aggression immediately after treatment compared to controls. At follow-up one year later, most of treatment effects had been maintained [
6]. In a more recent study, the same authors found that child training had a significant effect on children’s social competence that generalized from the school to the home setting [
7]. Similarly, Kazdin, Siegel and Bass conducted an RCT that evaluated Problem-Solving Skills Training and a parent-management training intervention and found the outcomes to be similar across settings (home, school and in the community) at post-treatment [
17]. One year later, the effects of the child training had been maintained compared to parent management training to a large degree.
Meta-analyses have indicated that social problem-solving training on its own primarily strengthens social-cognitive skills, compared to social skills training, which primarily improves social interaction skills [
18],[
19]; neither form of training influenced child problem behavior significantly. Schneider concluded in his review of 79 social skills programs that they generally had positive effects, but the outcomes were more positive when offered to withdrawn children than to unpopular or aggressive children [
20]. Younger children, immature children and aggressive children seemed to profit least from social skills training because either the contents or the presentation were insufficient or inadequate [
21]. In a systematic review, social competence training seemed to be most effective for children who had been exposed to critical life events and who lacked social stimulation, while children with externalizing and internalizing problems had approximately the same moderate outcomes [
19]. Children who were anxious, isolated and lonely seemed to benefit more than those conduct problems. A later meta-analysis demonstrated that social competence training had a small short-term effect on a broad range of behavioral and mental health problems and the long-term outcomes were smaller [
22]. In this review, the average post-test effect size (Cohen’s d) was .29 and the follow-up effect size was .21. These cognitive-behavioral programs turned out to produce the most sustainable outcomes for children with conduct problems, but the effect sizes were higher for social competence outcomes than for antisocial behavior outcomes. This review demonstrated that few studies assessed follow-up outcomes. Furthermore, the findings showed that most studies included small samples (70.8% had fewer than 49 participants), and only a few of the studies (10.4%) evaluated individually delivered social skills training. Kavale, Mathur, Forness, Rutherford & Quinn conducted a meta-analysis of 64 single-case studies of social skills training of children with social and emotional difficulties [
23]. The discouraging conclusion of this study was that individual social skills training had little support in research, even when it was frequently applied to this target group. In another meta-analysis, Quinn, Kavale, Mathur, Rutherford & Forness found a positive but small effect of individually delivered social skills training but concluded that such training on its own was not sufficient to prevent or reduce child conduct problems [
24]. An increase in social competence did not automatically result in reduction of conduct problems.
According to the literature, the greatest challenges to social competence training seem to be the lack of environmental support, which limits the generalization or transfer of training effects and lowers the sustainability of the program activity and outcomes [
21]. Formal monitoring of program implementation and control of intervention integrity is also often missing. It is often difficult to confirm whether a program has been delivered with the dosage and engagement that was intended by the program developer. The social validity (i.e., the degree to which an intervention is socially acceptable and/or relevant) of some programs has also been questioned, indicating that they apply to few of the skills that make up social competence.
In sum, social skills training has produced varying results in outcome studies, and children with conduct problems generally seem to benefit less than children with internalizing problems. The modest outcomes may be related to the limited content or duration of training, the mismatch of the intervention to the children’s needs, and the weakness of the evaluation designs (e.g., short time frame, lack of follow up studies, and few evaluations of the implementation quality).
Although the findings have been mixed [
25],[
26], there are indications of increased effects of social skills training when combined with behavioral parent training [
7],[
17],[
27]. When the promotion of cognitive and social competences is integrated with changing family interactions and increasing the quality of parenting, long-term protective effects have been demonstrated [
28]. While such studies are important, it is still necessary to examine whether social skills training has unique effects because previous evaluations have revealed mixed results, and it remains an open question whether social skills training adequately impacts conduct problems. To examine the unique effect of individually delivered social skills training on conduct problems, the current study evaluated the effectiveness of Individual Social Skills Training (ISST). Previous trials have largely been conducted within small samples efficacy trial designs and few trails have included a practice as usual comparison group. Thus, the current randomized effectiveness trial adds to the existing literature by including a relatively large sample, follow-up assessments, a practice as usual comparison group and measurement of intervention integrity.