Youth criminal justice systems are under growing pressure to reduce re-offending behaviour and support young people's health and developmental needs. This systematic review and meta-analysis sought to synthesise evidence for two prominent community-based interventions for delinquent and antisocial behaviour, multisystemic therapy (MST) and functional family therapy (FFT). We searched Medline, PsycInfo, Scopus, Web of Science, and Social Services Abstracts for randomised controlled trials (RCTs) and quasi-experimental studies evaluating MST/FFT. Included studies involved participants aged under 18; included interventions targeted delinquent/antisocial behaviour, but not maltreatment. We estimated effect sizes for 6 primary outcomes, synthesising RCTs comparing MST/FFT to usual care using correlated hierarchical effects meta-analysis. We assessed risk of bias and evidence strength using best-practice tools. Given the additional resources needed to implement MST/FFT, we rated evidence strength against a minimum clinically important difference rather than a null effect. This study is registered with PROSPERO, CRD42021279736. We included 35 studies for MST (16 RCTs meta-analysed comprising 4095 participants, 26% female) and 19 studies for FFT (7 RCTs meta-analysed comprising 1471 participants, 22% female). MST had a likely clinically important effect on time in out-of-home care, but no clinically important effects on other primary outcomes (delinquency, new offences/convictions, placement in out-of-home care, substance use), with low-to-moderate evidence strength. FFT demonstrated possible clinically important effects for the number of new offences/convictions, time in out-of-home care, and substance use, but evidence strength was low. Contrary to reports in some evidence clearinghouses indicating that MST/FFT are supported by the highest level of evidence strength, there is limited evidence that these interventions are superior to usual care in reducing delinquent and antisocial behaviour in adolescence. These findings should be viewed in the context of important methodological differences with prior reviews, including the rating of evidence strength against a minimum clinically important difference.
Read full abstract