Prevention of Internalizing Symptoms and Depression in the Familias Unidas Intervention: Examining Intervention Non-Responders

Introduction: An important question in intervention research is: For whom didn’t the intervention work?  Though the analysis of intervention main effects and moderation effects provide some answer to that question, a closer look at individuals for whom the intervention did not appear to work may be helpful in moving prevention science forward.  FamiliasUnidas is a family-focused prevention intervention targeting externalizing behaviors among Hispanic youth.  Though not designed to prevent youth depressive and anxiety symptoms, previous work has shown that FamiliasUnidas is effective in reducing internalizing symptoms among Hispanic adolescents.

Methods/ Results: This poster will use two distinct samples.  The first is a sample of 12- to 17-year-old adolescents (N=242) from the FamiliasUnidas- Juvenile Justice trial who were identified as having a history of delinquency. The second is a sample of 12- to 16-year-old adolescents (N=213) from the FamiliasUnidas- Targeted Referred trial who were referred to the intervention as a result of school counselor or parent reports indicating conduct, aggression and/or attention problems. The proportion of youth meeting the DISC DPS Major Depressive Disorder diagnosis cut-point at baseline was high, at approximately 21% in each sample. Causal inference modeling with growth mixture analysis will be used to identify and characterize clusters of individuals based on predicted response to both intervention and control conditions. Following the approach by Muthén and Brown (2009 Statistics in Medicine), these will be classified into responders for both intervention and control, responders to only one of these conditions, and nonresponders to either condition.  Predictors of class membership will be used to identify those who are nonresponsive to the intervention.

Conclusions: These analyses represent a further step towardimproving our understanding of adolescents for whom prevention interventions may not work. The use of two samples from the same intervention provides an opportunity to identify patterns of non-responders across similar studies which may inform future work in the area.