Category: Child / Adolescent - Depression

PS14- #A26 - Maternal Relationship Quality as a Mediator Between Childhood Maltreatment and Depressive Symptoms in Adolescence: Effects of Gender Moderation

Saturday, Nov 18
4:00 PM – 5:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Adolescent Depression | Gender | Mediation / Mediators

Childhood maltreatment has implications for the development of adolescent-mother relationships and symptoms of depression. Gender differences in the salience of interpersonal relationships and gender role identification in adolescence may lead females to reflect particularly poorly on their relationships with their mothers when they have a history of childhood maltreatment. The perceived poor quality of this relationship may then confer risk for later depressive symptoms, creating a unique developmental pathway for maltreated females. The present study examined gender differences in this developmental pathway from childhood maltreatment, to mother-adolescent relationships, to subsequent depressive symptoms in 342 (151 female, 191 male) maltreated (n = 198) and nonmaltreated (n = 144) youth. An observer report Q-Scale measure of depressive symptoms was developed and received preliminary support.


A confirmatory factor analysis was conducted using items from the California Q Sort selected by expert raters to represent a prototypical individual with depression. This model had good fit to the data, χ2 (8) = 18.662, p < .05, RMSEA = 0.058, CFI = 0.984, SRMR = 0.023, with factor loadings ranging from 0.504 to 0.855. This new Depressive Symptoms Q-Scale has the potential to make important contributions to assessment batteries as it increases the diversity of measures available to psychopathology researchers.


A multiple group structural equation model (SEM) was then tested to examine gender moderation. A partially unconstrained model allowed the paths from Wave 1 maltreatment to Wave 2 maternal relationship quality, and from Wave 2 maternal relationship quality to Wave 3 depressive symptoms, to be freely estimated across genders. This model had adequate fit to the data, χ2(154) = 242.314, p < .001; CFI = 0.925; RMSEA = 0.058; SRMR = 0.079, and a significantly better fit than the fully constrained model, Δχ2(2) = 6.686, p < .05. Maltreatment predicted significantly lower maternal relationship quality for females (β = -0.294, p < .001), but not for males (β = -0.069, ns). Moreover, lower maternal relationship quality predicted greater depressive symptoms for females (β = -0.203, p < .05), but not for males (β = -0.036, ns). Consistent with our hypotheses, lower maternal relationship quality significantly mediated the relationship between maltreatment and depressive symptoms for females (indirect effect estimate = 0.056, SE = 0.03, 95% asymmetric CI [0.006, 0.123]), but not for males (indirect effect estimate = -0.005, SE = 0.006, 95% asymmetric CI [-.019, .005]). Results suggest relational and attachment interventions may be particularly useful in the prevention and intervention of depression in adolescent females with a history of childhood maltreatment. 

Michelle Alto

University of Rochester
Rochester, New York

Elizabeth Handley

University of Rochester

Fred Rogosch

Mt. Hope Family Center, University of Rochester

Dante Cicchetti

University of Minnesota

Sheree Toth

Mt. Hope Family Center, University of Rochester