Linkages between childhood executive functioning and adolescent social functioning and psychopathology in girls with ADHD

Abstract

We followed an ethnically and socioeconomically diverse sample of preadolescent girls with ADHD (n = 140) and matched comparison girls (n = 88) over a period of 5 years, from middle childhood through early/midadolescence, with the aim of determining whether childhood levels of executive function (EF) would predict adolescent multi-informant outcomes of social functioning and psychopathology, including comorbidity between externalizing and internalizing symptomatology. Predictors were well-established measures of planning, response inhibition, and working memory, along with a control measure of fine motor control. Independent of ADHD versus comparison group status, (a) childhood planning and response inhibition predicted adolescent social functioning and (b) childhood planning predicted comorbid internalizing/externalizing disorders in adolescence. Subgroup status (ADHD-Combined, ADHD-Inattentive, and comparison) moderated the relationship between childhood planning and adolescent internalizing/externalizing comorbidity, with the combined type revealing particularly strong associations between baseline planning and adolescent comorbidity. Mediation analyses indicated that adolescent social functioning mediated the prediction from childhood EF to comorbidity at follow-up; in turn, in the girls with ADHD, adolescent comorbidity mediated the prediction from childhood EF to social functioning at follow-up. We conclude that childhood interventions should target EF impairments in addition to behavioral symptoms.

Figure 1

Regression lines for relations between the executive function of Planning during childhood and the presence of an internalizing disorder during adolescence as moderated by ADHD subgroup (ADHD-Combined, ADHD-Inattentive, and comparison), a 2-way interaction. b = unstandardized regression coefficient (i.e., simple slope); SD = standard deviation.

Figure 2

Regression lines for relations between the executive function of Planning during childhood and the presence of comorbidity between internalizing and externalizing disorders during adolescence as moderated by ADHD subgroup (ADHD-Combined, ADHD-Inattentive, and comparison), a 2-way interaction. b = unstandardized regression coefficient (i.e., simple slope); SD = standard deviation.

Figure 3

Mediated models for childhood planning, adolescent social functioning, and comorbidity between internalizing and externalizing disorders during adolescence. β coefficients for childhood planning are direct effects above the path and mediated effects below the path. * = p < .05; ** = p < .01, *** = p < .001.

Figure 4

Mediated models for childhood planning, adolescent comorbidity, and adolescent social functioning. β coefficients for childhood planning are direct effects above the path and mediated effects below the path. * = p < .05; ** = p < .01, *** = p < .001.