Social-ecological predictors of externalizing behavior trajectories in at-risk youth

Extant research consistently links youth externalizing problems and later maladaptive outcomes, and these behaviors are particularly detrimental given their relative stability across development. Although an array of risk and protective factors for externalizing problems have been identified, few studies have examined factors reflecting the multiple social-ecological levels that influence child development and used them to predict longitudinal trajectories of externalizing problems. The current study examined externalizing behavior trajectories in a sample of 1,094 at-risk youth (539 boys, 555 girls) from the Longitudinal Studies in Child Abuse and Neglect multisite longitudinal study of child maltreatment. Normed Child Behavior Checklist externalizing scores were used to estimate group trajectories via growth-based trajectory modeling at ages 10, 12, 14, and 16 using the SAS PROC TRAJ procedure. Model fit was assessed using the Bayes information criterion and the Akaike information criterion statistics. Analyses revealed optimal fit for five distinct behavioral trajectories: low stable, mid-increasing, mid-decreasing, medium high, and high stable. Multinomial logistic regressions revealed that a combination of risk and protective factors at individual, family, school, and neighborhood levels contribute to distinct trajectories of externalizing problems over time. Predictors of low and decreasing trajectories can inform interventions aimed at addressing externalizing problems among high-risk adolescents.