Research That Matters (January 17 - 20, 2008)

Current child welfare policy advocates for the placement of maltreated children into relative foster care arrangements when available. Studies have shown that these placements are more stable than non-relative out-of-home arrangements and that stable children are happier, healthier, and do better in school. However, it is not known if this policy promotes the stability of all children, particularly children with a disability. This study used data from the National Survey of Child and Adolescent Well-being (NSCAW), a nationally representative longitudinal sample of children who have come to the attention of child protective services (CPS). NSCAW contains both standardized tests that were used by the authors to construct objective indicators of disability (+2 std. dev. below mean) and questions relating to subjective determinations of disability by caregivers. The standardized tests include: measures of development (Battelle and Bayley Scales), language (Preschool Language Scale), functional independence (Vineland Adaptive Behavior Scale), educational achievement (Mini-Battery of Achievement), and behavior (Child Behavior Check List). Due to the age restrictions of the standardized tests, 4 separate models (ages 0-2, 3-5, 6-10, and 11-15) were developed using baseline indicators of a disability described above as well as the biological relatedness of caregiver to predict placement instability between September of 1999 and March of 2004. Instability, a dichotomous variable, was constructed from indicators of whether the interviewed caregiver at any of the 4 waves was the same caregiver in the preceding wave. At baseline, caregivers identified 1404 of the 5501 total sample of children as having a disability, or 30% of the weighted sample. Standardized tests indicated a wide variability of rates dependent on type of disability: 53% development, 17% language, 10% functional independence, 12% educational, and 43% behavioral disability. Bivariate analyses showed that more children with any type of disability were placed with non-relative out-of-home placements than with relative placements (33% vs 24%). Multivariate analyses, which also controlled for child and caregiver demographics and maltreatment type at baseline, showed that only a disability affecting functional independence by age 2 had significantly greater odds (OR = 6.06, p< .01) of instability. Interestingly, children with the same functional disability between the ages of 3 and 10 were significantly less likely to experience instability (OR = .10 and .19, p< .01) than all other children. Regardless of the type of disability and age, children placed with non-relative caregivers were consistently more likely to experience instability than children in relative placements. Interaction effects between placement and disability were entered into the models. Children under the age of 2 with a caregiver reported disability in relative placements were less likely to experience instability (OR = .12, p< .05) than all other children. Although most of the disability measures did not predict instability, being placed with a non-relative did consistently. The both negative (under 2) and positive (between 3 and 10) effect of a functional limitation necessitates further research on the caregiver-child relationship. Overall, this study speaks to the expansively positive effect of kinship on stability for all children.