Placement Instability, Neighborhood Context, and the Co-Development of Prosocial Skills and Mental Health Symptoms In Early Childhood

Background and Purpose: Children investigated by Child Protective Services are at elevated risk for developmental and mental health difficulties, and research suggests that children involved with the child welfare system experience particularly high rates of emotional and behavioral problems. Early childhood is also an important period for the consolidation of prosocial skills and the emergence of more stable internalizing and externalizing symptom patterns. Yet most existing research examines these domains separately and at discrete time points, limiting understanding of how they unfold together and how ecological contexts shape these interconnected trajectories. Guided by developmental cascade theory and ecological systems theory, this study examined the co-development of prosocial skills, internalizing symptoms, and externalizing behaviors among children ages 4–6 in the child welfare system and tested whether proximal factors (maltreatment type, placement instability) and distal factors (neighborhood environment quality) predicted baseline levels and rates of change across domains.

Methods: Data were drawn from the National Survey of Child and Adolescent Well-Being II (NSCAW-II), a nationally representative longitudinal study of children investigated for maltreatment (N = 708; ages 4–6 at baseline; 46% female; 43% White non-Hispanic, 32% Black non-Hispanic, 25% Hispanic). Outcomes assessed across three waves (baseline, 18 months, 36 months) included prosocial skills (Social Skills Rating System standard scores), and internalizing and externalizing symptoms (Child Behavior Checklist T-scores). Key predictors included caseworker-reported maltreatment type (physical, sexual, neglect), cumulative placement instability, and neighborhood environment quality. Parallel-process latent growth curve modeling was conducted in Mplus 8.11 using maximum likelihood estimation. Missing predictor data were addressed via Bayesian multiple imputation (40 datasets) pooled using Rubin's rules.

Results: The unconditional model demonstrated excellent fit (CFI = 1.00; TLI = 1.00; RMSEA = 0.00; SRMR = 0.007). Internalizing symptoms significantly decreased over time while externalizing symptoms significantly increased; prosocial skills remained stable on average but showed significant individual variability. In the conditional model, faster prosocial growth was associated with steeper declines in internalizing (r = −.55, p = .007) and externalizing symptoms (r = −.58, p = .001), while internalizing and externalizing changes were strongly positively correlated (r = .75, p < .001). Placement instability predicted lower baseline prosocial skills (β = −.22), higher baseline internalizing (β = .19) and externalizing (β = .17) symptoms (all p < .001), yet was associated with faster prosocial growth over time (β = .20, p = .007). Neighborhood environment quality predicted more favorable baseline functioning across all three domains. Physical maltreatment was specifically associated with elevated baseline externalizing symptoms (β = .10, p = .021).

Conclusions and Implications: Prosocial skills, internalizing symptoms, and externalizing behaviors appear to function as an interconnected developmental system, with gains in prosocial competence co-occurring with mental health improvement. Placement instability was the most consistent predictor across domains, underscoring the importance of placement stability as a practice priority while also suggesting the compensatory adaptation in prosocial development. Findings support interventions that simultaneously promote prosocial skill-building and reduce emotional and behavioral symptoms and point to placement and neighborhood context as important targets for intervention.