A Multilevel Meta-Analysis of Single-Case Research on Interventions for Externalizing Behavior Problems in Children and Adolescents
Externalizing interventions work great during treatment but fade fast unless you build ongoing parent or teacher support.
01Research in Context
What this study did
Kok and his team looked at 270 single-case studies.
All studies tested ways to cut hitting, yelling, and other acting-out in kids.
They used math that tracks both the child and the study level at once.
What they found
During treatment, every method worked.
Hitting and yelling dropped fast.
But six weeks later, most gains were gone.
Also, results changed more from study to study than from child to child.
Watching the child gave clearer proof than parent checklists.
How this fits with other research
Breider et al. (2024) and Farmer et al. (2012) show parent training keeps gains for six months in autistic kids.
Kok’s pool had few parent programs, so the fade-out may be about weak follow-up, not the kids.
Mammarella et al. (2022) found most school FBA studies skip real-world checks.
Kok’s fade-out could be linked to that same gap—tight lab setups don’t last in messy classrooms.
Stewart et al. (2018) meta-analysis saw small but lasting parent effects in autism.
Kok’s mixed-clinical sample shows the same small start, but shorter stay—hinting diagnosis may matter less than who keeps coaching the parents.
Why it matters
You can trust that your behavior plan will work while you run it.
To keep it working, add parent or teacher coaching that outlives direct sessions.
Also, film or count real acts instead of only using rating scales.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add one parent or teacher coaching call the week after you fade direct sessions to lock in gains.
02At a glance
03Original abstract
The overall effectiveness of interventions for youth externalizing behavior problems was studied using a review and a meta-analysis of published single-case research in children and adolescents. Scientific databases and gray literature were searched for quantitative single-case studies concerned with the treatment of externalizing behavior problems in children and adolescents. Study and case characteristics were extracted, and the studies were rated for quality. Raw graph data from individual cases were aggregated and analyzed by means of multilevel meta-analysis for single-case research. We identified 78 studies including 270 cases (mean age = 8.70 years; 71.48% male individuals). Overall, positive within-person changes during the treatment as opposed to baseline were observed. Reductions in symptoms did not carry through the follow-up phase. However, variations in treatment effects were observed, with larger variations among studies than among cases. Furthermore, studies using observational assessment methods yielded stronger results than studies using questionnaires to assess outcomes. Although the scores for the external validity of the studies were above average, the scores for internal validity were below average. Although part of the internal validity result can be attributed to underreporting certain quality standards in the studies, it is of great importance for the field of single-case research to start implementing existing methodological guidelines and to comprehensively report case-relevant information. This will, in addition, facilitate our understanding of the variability in treatment outcomes for specific children, and will enable us to learn more about the effects of interventions in diverse youth populations. Preregistration Meta-analysis of Single-Case Research on Interventions for Externalizing Behavioral Problems in Children and Adolescents; https://doi.org/10.17605/osf.io/4bewa
JAACAP Open, 2026 · doi:10.1016/j.jaacop.2025.12.002