Internalizing symptoms and conduct problems: Redundant, incremental, or interactive risk factors for adolescent substance use during the first year of high school?

Aim: The complex interplay of externalizing and internalizing problems in substance use risk is not well understood. This study tested whether the relationship of conduct problems and several internalizing disorders with future substance use is redundant, incremental, or interactive in adolescents. Methods: Two semiannual waves of data from the Happiness and Health Study were used, which included 3383 adolescents (M age = 14.1 years old; 53% females) in Los Angeles who were beginning high school at baseline. Logistic regression models tested the likelihood of past six-month alcohol, tobacco, marijuana, and any substance use at follow-up conditional on baseline conduct problems, symptoms of one of several internalizing disorders (i.e., Social Phobia and Major Depressive, Generalized Anxiety, Panic, and Obsessive-Compulsive Disorder), and their interaction adjusting for baseline use and other covariates. Findings: Conduct problems were a robust and consistent risk factor of each substance use outcome at follow-up. When adjusting for the internalizing-conduct comorbidity, depressive symptoms were the only internalizing problem whose risk for alcohol, tobacco, and any substance use was incremental to conduct problems. With the exception of social phobia, antagonistic interactive relationships between each internalizing disorder and conduct problems were found when predicting any substance use; internalizing symptoms was a more robust risk factor for substance use in teens with low (vs. high) conduct problems. Conclusions: Although internalizing and externalizing problems both generally increase risk of substance use, a closer look reveals important nuances in these risk pathways, particularly among teens with comorbid externalizing and internalizing problems. (C) 2016 Elsevier Ireland Ltd. All rights reserved.