Objective: The relationships between depressive symptoms and alcohol use among maltreated youth have been well documented. However, the direction of these associations remains unclear. Two prevalent but contrasting perspectives on these associations-the self-medication and impaired-functioning hypotheses-have each received extensive support. The present study aims to test these hypotheses and to reconcile these theoretical and empirical inconsistencies by advancing a third integrative conceptualization, the bidirectional hypothesis, which considers developmental timing in the transactional associations between maltreated youth's depressive symptoms and alcohol use over time.
Method: Data were drawn from the National Survey of Child and Adolescent Well-Being II, a longitudinal, nationally representative sample of youth involved with Child Protective Services because they have been maltreated. The current study included data from 657 youth followed over three equidistant time points for 3 years. Youth were aged 11-14 years at the first time point, with 52.7% female and 51.4% White.
Results: Autoregressive analyses within a structural equation modeling framework supported the impaired-functioning hypothesis during early adolescence and the self-medication hypothesis during mid- to late adolescence. The data were consistent with the bidirectional hypothesis in which depressive symptoms and alcohol use differentially predicted each other depending on the stage of adolescence.
Conclusions: The bidirectional associations between alcohol use and depressive symptoms are influenced by developmental timing. Researchers should consider age when tailoring prevention programs that target maltreated youth. Depressive symptoms should be screened early and frequently to prevent adolescent substance use.