Date of Award

January 2011

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)



First Advisor

Elena L. Grigorenko

Subject Area(s)

Psychology, Developmental psychology


Despite the enormous cost of the foster care system, high rates of psychopathology and homelessness among young-adult foster care alumni provide a stark reminder of the challenges faced by this vulnerable population. This study characterizes the effect of a history of foster care on psychopathology in a group of 39 adolescents that had exited foster care and were reunified with their biological mothers. A history of foster care was defined as out-of-home placement by child welfare for at least one month; median foster stay was 1.5 years and median age at placement was 8.5 years. A control group of 78 adolescents was matched with the foster group using exact and logistic-regression nearest-neighbor methods. Matched variables included well-established, major childhood risk factors for the development of psychopathology: maternal substance abuse, maternal psychopathology, and childhood maltreatment (i.e. physical and sexual abuse, neglect and domestic violence). With the two groups matched in this way, recent research suggests that the two groups had comparable histories of adverse childhood events, and thus the major inter-group difference is a temporary separation from the biological mother, enforced by child welfare services (i.e foster care). Participants, and their mothers as second reporters, completed self-report, parent-report and structured interview assessments, providing data on major psychiatric diagnoses and symptom scales. The prevalence of externalizing diagnoses in the foster group was 41.0% (24.9% - 57.2%) compared with 19.2% (10.3% - 28.2%) in the control group (r = .25). Substance dependence prevalence was 25.6% (11.3% - 40.0%) compared with 5.1% (0.1% - 10.1%) in the control group (r = .30). The foster group also had more depression symptoms as measured by three assessments (p < .05, r = .21 to .25); the foster group also had more overall externalizing symptoms (p = .015, r = .22), including conduct problems (p = .007, r = .25) and hyperactivity (p = .023, r = .21). For every comparison made, the foster group demonstrated more psychopathology than controls. Thus despite the protective goal of foster care, it may have detrimental effects on the child's subsequent development. Further studies with larger sample sizes are needed to confirm these findings that have the potential to substantially alter child welfare policy by reducing the number of foster care placements in favor of other child and family support services.


This is an Open Access Thesis.

Open Access

This Article is Open Access