Poster #36 - A Family-Level Typology of Depressive Symptoms: Longitudinal Relations to Adolescent Risk-Taking

• In Event: Poster Session 04

Abstract

Although some risk-taking during adolescence is expected, risky behaviors can have consequences for adolescent adjustment, such as lower academic achievement, alcohol dependence and substance abuse, teenage pregnancy, and legal problems (Maggs et al., 1997; Marshall, 2014). It is important to understand predictors of greater risk-taking to curb threats to adolescents’ safety and psychosocial well-being. Adolescent depressive symptoms and impaired family functioning relate to more risk-taking behaviors (Hooshmand et al., 2012; Lorenzo-Blanco et al., 2017). However, research has yet to use a holistic view of how depressive symptoms across multiple family members at salient developmental periods are longitudinally related to adolescents’ later risk-taking. Using the NICHD Study of Early Child Care and Youth Development, we investigated how depressive symptoms across the family during a transitional period (fifth grade) related to risk-taking when children reached age 15.
Participants included 651 two-parent families. Half the children were females (50%), 84.9% were white/non-Hispanic, 6.4% Hispanic, 5.3% African American, and 3.3% other/multiple ethnicities. Half of families were middle-income according to income-to-needs ratios. Mothers and fathers reported on their depressive symptoms using the Center for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977). Children completed the Child Depression Inventory (Short Form; CDI-S; Kovacs, 1992) and the Risky Behavior Questionnaire.
We utilized a latent profile analysis to classify families into groups based on mothers’, fathers’, and children’s levels of depressive symptoms at fifth grade. We tested models with 2 to 6 profiles and compared fit indices to determine the best-fitting solution. The 5-profile solution was chosen, as it was a significant improvement on the 4-profile solution and showed clear delineation between the groups (see Table 1). Five groups emerged (see Figure 1): one in which mothers reported very high symptoms and children reported moderately high symptoms (Very High Mom, Moderately High Child; 3.7% of the sample), one in which only mothers reported very high symptoms (Very High Mom; 10.6%), one in which only children reported very high symptoms (Very High Child; 6.6%), one in which children reported moderately high symptoms (Moderately High Child; 16.4%), and a final group in which all three family members reported low levels of depressive symptoms (All Low; 62.7%).
An analysis of covariance (ANCOVA) was used to determine differences in adolescent risk-taking between the profiles, controlling for each family member’s concurrent depressive symptoms at age 15, child gender, and family income. The ANCOVA revealed significant group differences in risk-taking, F(4, 641)=5.724, MSE=20.651, p<.001. Post-hoc comparisons using a Bonferroni adjustment showed that membership in the Very High Mom, Moderately High Child profile predicted significantly greater adolescent risk-taking at age 15 compared to all other groups.
Interestingly, adolescent risk-taking was only predicted by membership in the one group in which both mother and child reported depressive symptoms. This suggests that child depressive symptoms alone may not be as predictive of problematic adolescent behavior; the co-existence of such internalizing symptoms in the family during a salient transitional period marked by pubertal and school changes may be necessary to significantly impact adolescents’ psychosocial outcomes.

Authors

• Mariam Hafiz, University of Texas at Dallas
  Presenting Author

• Jackie A Nelson, University of Texas at Dallas