Adaptive Intervention Strategies in Conduct Problems Prevention

This project proposes to conduct feasibility research to inform implementation of a future full-scale SMART design (i.e., sequential, multiple assignment, randomized trial) that will be used to construct adaptive intervention strategies (AIS) for conduct problems prevention. AIS individualize treatment via decision rules that specify how the type (youth-focused or parent focused) or intensity (low dosage or high dosage) of an intervention should be formulated prior to the beginning of treatment based on youth and family characteristics and/or repeatedly adjusted over time based on proximal outcomes collected during treatment. AIS are needed in conduct problems prevention to address the heterogeneity of at-risk youth and the variability in response to conventional fixed-type preventive interventions. With the present SMART trial each participant will progress through two stages of intervention using a stepped-care framework. In the first stage participants will be randomized to one of two 'brief-type' intervention options, either the youth-focused Teen Intervene-Brief program (TI-B; Winters & Leitten, 2007) or the parent-focused Everyday Parenting-Brief program (EP-B; Dishion et al., 2003, 2010). Responders to either program will be stepped down and monitored over time for maintenance. Non-responders to either program will be stepped up and randomized to one of two second stage 'intensive-type' intervention options that feature either (1) continuation of the first stage option with increased dosage (EP-Expanded or TI-Expanded), or (2) switching to the alternative expanded intervention modality. This feasibility study will enroll high risk youth (10-17 years of age) who have been arrested for status or misdemeanor offenses and referred for pre-court juvenile diversion programming. The aims of this feasibility research are to (1) develop practice infrastructure for implementing a SMART design and assess practitioner adherence to the various intervention sequences, (2) roll out the stepped-care intervention sequences and obtain estimates of recruitment into SMART, attrition at both stages, and overall response rate to first-stage intervention options, (3) describe the demographic and clinical characteristics of the sample of diversion-referred youth who are enrolled in the study, (4) create a latent construct for conduct problems that will be used as the distal outcome, and (5) explore the utility of incorporating secondary tailoring variables (e.g., child and family risk characteristics) in the adaptive intervention model.