Individual Submission Summary
Share...

Direct link:

Poster #48 - Anxiety and Callous Unemotional Traits as Predictors of Conduct Disorder in Middle Childhood

Thu, March 21, 12:30 to 1:45pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Introduction
Children from low income, urban neighborhoods are susceptible to early onset development of behavioral problems (Lahey & Loeber, 1994; Moffitt, 1993; Merikangas et al., 2010). Among these disruptive behaviors is Conduct Disorder (CD). The literature is mixed regarding the role anxiety disorders. Generalized anxiety (GAD) was found to increase risks of conduct problems (Moffit et al., 2007). In contrast, Walker et al., (1991) thought of anxiety as a protective factor against CD due to low levels of maladjustment. Children of color, specifically Black children, are more likely to have Callous Unemotional (CU) traits (e.g. lack of empathy and guilt for their misbehavior, antisocial traits). Research consistently conceptualizes CU traits as having low anxiety and higher CU traits, making them negatively associated (Frick et all., 1999; Pardini et al., 2007; Frick & White, 2008 ).
The aim of this study is to see if anxiety and CU traits predict CD, in addition to testing if the interaction between anxiety and CU traits predict to the behavioral disorders to see if they co-occur.

Methods
The total sample consists of 104 children (M = 9.93, SD = 1.22, 50% male; 95% African American, 5% Latino) from the Philadelphia, PA public school district, as well as reports from primary caregivers and teachers. There are self-report measures on anxiety (separation, general, and panic), CU traits, and CD.
It is hypothesized that children with high CU traits will be more at-risk for acquiring CD, with anxiety having only impacting those with existing CU traits.

Results
Regression analyses included age, sex, GAD symptoms (rated by all informants), parent-rated CU behaviors, and the GAD×CU interaction term with CD symptoms (rated by all informants) as the outcome. Two significant interactions emerged: teacher-rated GAD×CU predicted parent-rated CD symptoms (β = .18, p = .05) and youth-rated GAD×CU predicted youth-rated CD symptoms (β = .22, p = .02). Post hoc probing indicated that the simple slope was significant among youth with higher (Bs = .41 and .52, respectively; both ps < .01) but not lower (Bs = -.08 and .16, respectively; ps > .19) levels of CU behaviors (Figures 1 and 2). Findings indicate that youth with lower levels of GAD symptoms experienced similar levels of CD symptoms regardless of their levels of CU behaviors. Youth with higher levels of GAD symptoms were rated as experiencing higher levels of CD symptoms in the context of higher levels of CU behaviors. Thus, the combination of higher levels of CU and higher levels of GAD symptoms was associated with the highest levels of CD symptoms, though these findings differed based on informant.

Significance
This study shows that CU traits are positively associated with high GAD symptoms. This is counters existing research stating that they are negatively associated (Frick et all., 1999; Pardini et al., 2007; Frick & White, 2008). The use of multiple informants is important in this study because it shows the proximity each reporter has to the symptoms, with the child being the closest and the teacher being the most distant.

Authors