DO INTERPERSONAL VARIABLES MODERATE OUTCOMES IN A RANDOMIZED DEPRESSION PREVENTION TRIAL? Do interpersonal variables moderate outcomes in a randomized depression prevention trial?
Interpersonal Psychotherapy-Adolescent Skills Training (IPT-AST) is an effective depression prevention program for youth with elevated symptoms of depression. The current study investigated interpersonal variables as moderators and predictors of depression prevention program outcomes to answer the important question of for whom these interventions were more or less effective. Youth were randomized to either IPT-AST (N = 95) or Group Counseling (GC) (N = 91), an active and stringent control
... led by school counselors, as part of the Depression Prevention Initiative (DPI), the largest study of IPT-AST to date. Adolescent depressive symptoms and global functioning outcomes were measured from baseline through 6-month follow-up. At baseline, adolescents reported on family functioning, as well as conflict and support in three relationships (mother, father, and sibling), which were averaged to examine family conflict and family support. Conflict and support in individual relationships were also examined. As hypothesized, a marginal moderation effect emerged for family conflict on rates of change in depressive symptoms. At low and average levels of conflict, youth in IPT-AST reported significantly greater rates of change than youth in GC. However, the ! ! iii! intervention difference dissipated at high levels of conflict. This moderation effect appeared to be driven by father and sibling conflict, which were both significant moderators of change in depressive symptoms. Mother conflict, while not a significant moderator, emerged as a marginal predictor of rates of improvement in depressive symptoms. As baseline mother conflict increased, adolescents across both interventions experienced greater reductions in depressive symptoms. Father support emerged as a marginally significant moderator of global functioning, and father conflict emerged as a significant predictor of global functioning. Effects of support and conflict in each relationship are outlined in detail in the paper, and implications of the findings are discussed. There was some evidence that at low and average levels of conflict, IPT-AST may have unique gains over GC, but at high levels of conflict, a more intensive intervention may be necessary. Further exploration and replication of moderation effects could help move the field forward toward personalized prevention. ! ! iv! Acknowledgements