| The decision making process in child welfare agencies concerning drug-exposed newborn infants is a complex and controversial issue throughout the United States. Drug-exposed newborns, as well as toddlers and preschool children of substance-abusing parents, represent the fastest growing segment of the foster care population, threatening to overwhelm an already stressed system. While researchers have investigated the factors that predict foster care placement for maltreated children, little is known about how the decision is made for drug-exposed newborn infants.; The objective of this study was to identify factors that predicted the placement status of 291 drug-exposed newborn infants referred to Child Protective Services in the Baltimore City Department of Social Services between July 1, 2000 and December 31, 2001. Two dependent variables were created: placement with parent(s) versus out-of-home placement and placement with parent(s) or relatives versus foster care. The mother's CPS case record was reviewed to extract data on 24 variables in four domains pertaining to the infant, the mother, the family, and the environment.; Crude odds ratios were calculated, and logistic regression analyses were conducted to identify the effect of the predictors on the likelihood of placement. Results showed that African American infants who exhibited withdrawal symptoms or drug effects were more likely to experience out-of-home placement. Infants who had a negative toxicology screen and whose father was available and appropriate to provide care were less likely to experience placement. Logistic regression results for placement with parent(s) or relatives or placement in foster care showed that infants whose father and relatives were available and appropriate were less likely to experience foster care.; A z statistic was computed to compare the logits of identical predictors across analyses for the two dependent variables. A comparison of the findings from the two models identified one predictor that differed significantly---availability of relatives. In addition to this difference, sensitivity (94.6%) and specificity (92.4%) findings suggest that the foster care model has greater predictive efficiency. Placement decisions for drug-exposed newborn infants are consistent with existing statutory law. |
