Testing Competing Models of Temporal Ordering between Family Conflict, Adolescent and Maternal Depression in Urban Families: Differences By Sex

Background/purpose: Maternal and adolescent depression are social problems which often co-occur in urban families. Family conflict may be a catalyst, but less is known about the temporal ordering between adolescent and maternal depression in relation to family conflict. For example, while family conflict could precede depression in adolescents and/or mothers, it may be that adolescent and/or maternal depression precedes family conflict. Alternatively, recent evidence also suggests reciprocal effects between family conflict and depression may explain these relationships. Further, researchers have found these relationships could differ by sex. For instance, adolescent females were found to be at greater risk for subsequent/recurrent depression if they reported higher family conflict, whereas adolescent males were not. In order to better understand these relationships, our study set out two aims: 1) evaluate three competing theoretical models (i.e., symptom-driven, interpersonal risk, and transactional) to understand predictive associations between family conflict, maternal and adolescent depression in urban families, and 2) examine these relationships for differences by youths’ sex.

Methods: Three time-points from the Young Adolescent Project that included 187 urban mother-adolescent dyads (93 males/94 females) were fit to autoregressive cross-lagged structural equation models to simultaneously evaluate competing theoretical models. Youth and mothers each self-reported depression and family conflict; youth- and mother-reported family conflict were analyzed in separate models. Covariates included youths’ age and maternal education. Multiple-group models examined differences by sex using equality constraints and the nested chi-square 1 df difference test.

Results: Results indicated a symptom-driven model whereby increases in adolescent depression predicted increases in youth-reported family conflict (b=.17 and b=.16, T2 and T3 respectively, both p<.01). For males, increases in adolescent depression predicted increases in maternal depression (b=.25/.23, p=.001), and increases in maternal depression predicted decreases in youth-reported family conflict (b=-.13/-.16, p<.05). For females, increases in adolescent depression also predicted increases in youth-reported family conflict (b=.18/.15, p<.05), but not maternal depression; instead, for females, increases in maternal depression predicted increases in mother-reported family conflict (b=.14/.15, p<.05). Analyses using parent-reported family conflict found zero significant pathways.

Conclusion/implications: Our findings highlight the importance of addressing adolescent depression in order to prevent subsequent family conflicts for urban youth, both for males and females. In our study, family conflict was not associated with subsequent depression as previous studies have found. Instead, for males, depression predicted subsequent increases in maternal depression, which then predicted decreases in youth-reported family conflict – possibly indicating maternal disengagement/withdrawal. However, for females, increases in maternal depression predicted increases in mother-reported family conflict. These results suggest distinctive targets for the prevention/intervention of family conflict based on adolescents’ sex and adds to prior knowledge on the relationship between maternal depression, female adolescent depression, and family conflict. Beyond sex differences, given previous research on adolescent depression and fluctuating degrees of family conflict, future research may want to investigate whether these effects vary by developmental stage (e.g., early, mid, and late adolescence) using an accelerated longitudinal cohort design, which could further our knowledge about these relationships and reveal possible time-sensitive periods when intervention/prevention efforts may be the most effective.