This purpose of this study was to determine whether it was possible to identify casework activities after child protection substantiation for neglect that would predict greater likelihood of foster care entry for African American children in four metropolitan mid-west counties.

The hypotheses were that service activities would vary by race and that this variation would be associated with differential entry into foster care after case substantiation.

This was a retrospectively conducted longitudinal case record reading study following the children from the time of substantiation through at least one year post substantiation. The subjects were 206 African American and Caucasian children matched on age group (0-5, 6-11), gender, county and reason for substantiation (specific type of neglect such as abandonment, failure to provide, lack of supervision).

The two step analysis was conducted using McNemar chi square to detect the significance of any observed differences between members of pairs on major case decisions and outcomes such as the decision to provide child protection services, case closure, placement and reunification. Stata's conditional logistic regression was used to identify variables predictive of case decisions when controlling for pair membership. In addition, interactions were run with race on all hypothesized predicting variables, such as mother's problems, history of child abuse and child's age.

There were no significant differences between African American and Caucasian children on case services, case decisions or case outcomes when using the McNemar and the conditional logistic regression. However, when the variables were run in interaction with race, there were significant interactions between race and age, race and Mother's drug problems, race and history of maltreatment and race and Mother's financial problems. For example, Caucasian children were more likely to go to placement after substantiation if they were one year old or younger while African American children were more likely to go into placement if they were ages five through nine. These relationships were significant at the .01 level.