Study suggests long-term foster care does not worsen children’s behavior problems over time

Date Published: 

Bellamy, J. (2008). Behavioral problems following reunification of children in long-term foster care. Children and Youth Services Review, 30, 216-228.

Reviewed by: 
Elizabeth Fast

A small number of studies have examined mental health outcomes for children placed in long-term foster care. The few studies that have been conducted investigated only limited populations in close geographic proximity. This secondary data analysis used a sample of 604 children from the National Study of Child and Adolescent Well-being (NSCAW) who had been in foster care for at least 12 months. Children who were reunified with their parents were compared to those who remained in foster care with respect to internalizing and externalizing behavior problems using the Child Behavior Checklist (which was completed by the current caregiver) both at the beginning of the study and after 36 months. Children who remained in foster care were found to have more behavioral problems, overall; however after 36 months, a decrease in problems was observed for these children. Children who were reunified with their families between baseline and 18 or 36 months follow-up and had not returned to care during the 36 months had a sharp increase in internalizing problems after an 18 month period. Once again, however, these problems had decreased after 36 months, and it should be noted that more than twice as many reunified children showed clinical levels of internalizing behavior problems by the end of the study period. The analyses performed suggest that the increase in internalizing problems for reunified children (and a somewhat stable pattern of externalizing problems) is probably associated with long-term family difficulties rather than any direct effect of the reunification process.

Methodological notes: 

This was a secondary data analysis of a sub-sample of 604 children from the NSCAW database and was completed using a two-stage combined stratification and cluster design. The study inclusion requirements for the long-term foster care sample consisted of: 1) children having been in care for at least 12 months at the time of sampling; 2) children were in placement at the time of sampling; 3) placement was preceded by a child welfare investigation. One limitation of this study is that these mental health trajectories only apply to children who spent at least one year in foster care and the majority of children who come into care do not fall into this category. Therefore, these data tell us nothing about outcomes for children who spend less than one year in foster care. Descriptive statistics, logistic regression and propensity score matching were used to measure the role of risks in the family context and children’s well-being in three waves: at baseline, 18 and 36 months. Children under two years of age were not included due to a lack of appropriate tools for measuring behavioral problems in very young children. Multiple Imputation (MI) was used to account for missing data.