Search
Program Calendar
Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Search Tips
Virtual Exhibit Hall
Personal Schedule
Sign In
X (Twitter)
Childhood conduct problems (CPs) and attention-deficit/hyperactivity disorder (ADHD) present an important public health and safety concern, as they are independently linked with impairments across several functional domains, including social competence, school readiness, and achievement (Chen et al., 2015; Kieling et al., 2010). Indeed, individuals who engage in life-course persistent antisocial behavior (5-10% of the population) account for over half of all crime in the United States (Eme, 2015). Independent investigations of childhood ADHD and CPs have delineated stable low, decreasing (a.k.a., childhood-limited), increasing (a.k.a., adolescent-onset/limited), and stable high (e.g., life-course persistent) behavioral trajectories (e.g., Moffitt et al., 2002; Sasser et al., 2015). However, these disorders are often comorbid and may have shared etiology (Beauchaine & Gatze-Kopp, 2012). Nevertheless, significant clinical heterogeneity in externalizing behavior demonstrates the importance of considering not only ADHD-CP comorbidity, but also other potential differentiating factors, such as callous-unemotional (CU) behaviors (Rowe et al., 2010). The current study used latent class analysis (LCA) to examine longitudinal parent- and teacher-rated trajectories of ADHD, CPs, and limited prosocial behavior (as a proxy for CU behaviors) from preschool to 5th grade.
Participants were 1155 families in a birth cohort study of children living in non-metropolitan areas of high poverty in the United States. Primary caregivers rated children’s hyperactivity, CPs, and prosocial behavior using their respective subscales on Strengths and Difficulties Questionnaire (SDQ) at 48, 58, and 90 months old. Teachers rated the same behavior yearly from childcare/preschool (3 years old) to 5th grade. Using hyperactivity and CP normed cutoffs, children were classified as either at/above (elevated) or below the 90th percentile (not elevated); a similar 10th percentile cutoff indexed limited prosocial behavior. A latent class analysis (LCA) examined whether children exhibited discrete trajectories of these pathological behaviors across childhood.
A four-class solution was the best-fitting model, BIC=22342.58, ssBIC=21951.89, aVLMR=447.94, p=.049. The largest class (47.5%) delineated a low-stable trajectory with children showing a low probability of any pathology when rated by both parents and teachers across all time points. The second largest class (22.3%) indexed moderately high probability of elevated of hyperactivity and CPs over time when rated by parents, but consistently low probability according to teachers’ ratings, with both parents and teachers rating children as having low probability of limited prosocial behavior. A third class (17.3%) indexed low probability of any pathology when rated by parents, but increasing pathology when rated by teachers. The smallest class (13.0%) indexed high stable probability of hyperactivity and CPs by both parents and teachers—however, parents consistently rated the probability of limited prosocial behavior as moderately low and teachers rated it as high.
We found multiple differentiated trajectories of corresponding ADHD symptoms, CPs, and limited prosocial behavior, including a group of children showing consistently high pathology across domains, who are likely at the highest risk for multi-domain impairments. Interestingly, we also found that ratings differed across parents and teachers for some children, demonstrating the importance of considering the context of behavior for understanding psychopathology. Implications and future directions will be discussed.
The Family Life Project Key Investigators