Family-based treatments for serious juvenile offenders : a multilevel meta-analysis

Abstract

Violent criminal acts and other serious crimes committed by youths result in considerable emotional, physical, and economic effects on these youths, their families, crime victims, and the larger community. Researchers have identified several family-based treatments that hold considerable promise in reducing serious juvenile offending; however, these treatments remain underutilized by youth service systems. In the present study, I used multilevel meta-analysis to (a) characterize the nature and quality of research on family-based treatments for serious juvenile offenders, (b) summarize the findings of this research by estimating an average effect size, and (c) examine the influence of moderators (e.g., characteristics of participants, treatments, and methods) on treatment outcomes. Results from 31 studies revealed a number of methodological strengths including frequent use of random assignment and comparison to usual treatment conditions. In addition, the meta-analysis synthesized 325 effect sizes from 29 of these studies and revealed that family-based treatments produced small but meaningful treatment effects (mean d = 0.26) relative to comparison conditions. Furthermore, moderator analyses revealed that certain characteristics of participants, study methods, and measures influenced the size of treatment effects; for example, effect sizes varied by domain of outcome measure such that measures of substance use showed the largest treatment effects and measures of peer relationships showed the smallest. Overall, the results of the present study have implications for policymakers, administrators, and treatment providers who make decisions about the dissemination and implementation of family-based treatments for serious juvenile offenders. In addition, researchers who seek to develop and study these treatments may wish to consider the current findings.