Purpose: Group therapy is the most popular form of treatment for substance abuse problems and there are many reports of its beneficial effects. Yet some studies have found that aggregating youth with conduct disorder (CD) created environments in which youth exhibited negative behavior that was positively reinforced by other group members (i.e., "deviancy training"). Other studies have suggested that antisocial behavior and deviancy training also were affected by group leader behaviors. These findings suggest that while CD group composition, member-reinforced disruptive behavior in groups, and leadership behavior may have independent effects on outcomes, they likely interact to produce diverse outcomes. This presentation reports on a NIAAA-funded study that examines the influence of CD group composition (i.e., proportion of CD among group members), disruptive behavior in group (i.e., antisocial behavior and deviancy training), and group leader behaviors (i.e., negative and positive) on changes in AOD use over time. The study tests whether the interactions of group leader behaviors with group disruptive behavior mediate changes in the trajectory of AOD behaviors over sessions and through follow-up.

Methods: Data derive from a completed NIAAA clinical trial involving 122 youths randomly assigned to 21 groups using the Westchester Model Student Assistance Program protocol. In this new study, evaluators reviewed audio tapes and transcripts of group sessions to rate group leader behaviors and disruptive behavior.

Results: Preliminary data have identified clear examples of sequencing of antisocial behavior, its reinforcement, and positive and negative leader behaviors in response to disruptive behavior. Findings suggest that disruptive behavior occurs in even “well-behaved” groups, and its development seems to rely on particular leader behaviors.

Discussion/Implications: Group work researchers should examine the conditions, circumstances, and mechanisms of actions within AOD groups that make them helpful and harmful. This project examines the role of three empirically-derived contributors to response to group treatment among adolescents, and tests a model about how they independently and interactively contribute to outcomes. Recommendations for composition and leadership behaviors are made to improve the efficacy of group work with adolescents with substance abuse problems.