Making 'what works' work: A meta-analytic study of the effect of treatment integrity on outcomes of evidence-based interventions for juveniles with antisocial behavior

This study meta-analytically examined the effect of treatment integrity on client outcomes of evidence-based interventions for juveniles with antisocial behavior. A total of 17 studies, from which 91 effect sizes could be retrieved, were included in the present 3-level meta-analysis. All included studies, to a certain level, adequately implemented procedures to establish, assess, evaluate and report the level of treatment integrity. A moderator analysis revealed that a medium-to-large effect of evidence-based interventions was found when the level of treatment integrity was high (d = 0.633, p < 0.001), whereas no significant effect was found when integrity was low (d = 0.143, ns). Treatment integrity was significantly associated with effect size even when adjusted for other significant moderators, indicating the specific contribution of high levels of treatment integrity to positive client outcomes. This implies that delivering interventions with high treatment integrity to youth with antisocial behavior is vital.