Theories of Change and Mediators of Psychotherapy Effectiveness in Adolescents With Externalising Behaviours: A Systematic Review

Background: Externalising behaviours are becoming a remarkably prevalent problem during adolescence, often precipitating both externalising and internalising disorders in later adulthood. Psychological treatments aim to increase the social functioning of adolescents in order for them to live a more balanced life and prevent these negative trajectories. However, little is known of the intervening variables and mediators involved in these treatments' change mechanisms. We conducted a systematic review, exploring the available evidence on mediators of psychological treatments for externalising behaviours and symptoms amongst adolescents (10 to 19 years old).

Methods: A systematic search was performed on Medline and PsycINFO databases, which identified studies from inception to February 23, 2020. Eligible studies included randomised controlled trials that enrolled adolescents with externalising symptoms and behaviours as, at least, one of the primary outcomes. A group of 20 reviewers from the COST-Action TREATme (CA16102) were divided into 10 pairs. Each pair independently screened studies for inclusion, extracted information from the included studies, and assessed the methodological quality of the included studies and the requirements for mediators, following Kazdin's criteria. Risk of bias of RCTs was assessed by the Mixed Methods Appraisal Tool. Extracted data from the included studies were reported using a narrative synthesis.

Results: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA), after removing duplicates, 3,660 articles were screened. Disagreements were resolved by consensus. In a second stage, 965 full-text articles were assessed for eligibility. A total of 14 studies fulfilled all inclusion criteria. The majority were related to systemic psychological treatment approaches. Two types of mediators were identified as potentially being involved in the mechanisms of change for better social improvements of adolescents: to increase healthier parent-adolescent relationships and parental discipline. However, there were significant and non-significant results amongst the same mediators, which led to discussing the results tentatively.

Conclusions: Family variables were found to be the largest group of investigated mediators, followed by relational, behavioural, and emotional variables. No cognitive or treatment-specific mediators were identified. Both adequate behavioural control of adolescents' peer behaviour and a better positive balance in their relationships with their parents seemed to buffer the effects of externalising behaviours in adolescents. Several methodological limitations concerning mediation testing design, outcome measures, and mediator selection have been identified.

Ethics and dissemination: Ethical approval was not required. PROSPERO registration number: CRD42021231835.