Preventing Post-Adoption Instability: Latent Classes and the Mediating Role of Parent–Child Relationships

Background and Purpose: Children adopted from U.S. foster care often carry trauma stemming from early adversities such as abuse, neglect, and placement disruption. Although adoptive parents intend to provide a supportive home, they may be unprepared for these compounding challenges, and adoption instability, both formal (e.g., foster care reentry) and informal (e.g., running away), may follow. Ensuring that every child has the opportunity to thrive in a permanent family is a foundational commitment of a just and inclusive society, and one that depends on rigorous evidence about what sustains adoption. Guided by the Post-Adoption and Guardianship Instability Conceptual Framework, this study identifies latent foster care adoption classes and tests whether parent–child relationship quality mediates the association between early adoption experiences and post-adoption instability, to inform prevention efforts for the target population.

Methods: Drawing on the National Survey of Child and Adolescent Well-Being (NSCAW) Adoption Follow-up Data (N = 295), this study first performed latent class analysis using six indicators: child behavior problems (measured by Child Behavior Checklist [CBCL]), neglect history, long foster care stay, foster care placement stability, parent preparedness, and parent motivation (i.e., want to help a child in need). This study further applied a generalized structural equation model to test whether parent–child relationship quality, measured by the Nurturing and Attachment subscale, mediated associations between latent adoption classes and post-adoption instability, while adjusting for child, caregiver, and adoption characteristics. IRB approval was obtained from [blinded for review].

Results: Three distinct classes emerged: Class 1, low-risk exposure and prepared caregivers; Class 2, unstable and prolonged foster care stay with unprepared caregivers; and Class 3, child behavioral problems. Approximately 20% of Class 1, 30% of Class 2, and 48% of Class 3 experienced post-adoption instability. Compared to Class 1, Classes 2 and 3 had poorer parent–child relationships (b₂ = −0.47, p < .001; b₃ = −0.57, p < .01), and stronger relationship quality was associated with lower adoption instability (bₘ = −0.53, p < .001). Relationship quality fully mediated the link between class membership and instability for both Class 2 (b₂ × bₘ = 0.25, p < .05) and Class 3 (b₃ × bₘ = 0.30, p < .05).

Conclusions and Implications: Findings from this study offer a nuanced view of pathways to adoption stability and point to relational health as a leverage point for prevention. Post-adoption services should prioritize attachment-focused support, as strong parent–child bonds can buffer instability risks tied to a child's trauma history or behavioral challenges. More broadly, this work illustrates how applied social work research can inform policies that strengthen families and protect the stability and belonging of children in the adoption system, advancing the democratic ideal that no child should be left without a family equipped to keep them.