Purpose: Research indicates that child characteristics can have a profound impact on the achievement of important child welfare outcomes such as those identified as Child and Family Service Reviews indicators. However, evaluation of the contribution of child characteristics to such outcomes has traditionally been a challenging undertaking due to the paucity of reliable data available in administrative databases. Moreover, when child characteristics including health or disability are taken into consideration, these constructs generally are not disaggregated into their physical health, mental health and developmental disability aspects. This study is an effort to evaluate the impact of child characteristics, including demographic, placement, health, mental health, and developmental disability characteristics, on the Federal outcome indicators of reunification, adoption, and placement stability. Methods: The Illinois Study of Child Well-being incorporated interviews with caregivers and caseworkers of children in care. Medical record reviews were carried out by trained nurses, and educational record reviews were carried out by teachers. Complete data concerning outcomes were derived from the DCFS administrative database. The sample was a random sample stratified by age and time in care. The sample was appropriately weighted to account for probability of selection and for non-response. The sample was comprised of 351 children. Analyses were conducted using SUDAAN. All reunification, adoption, and stability outcomes since the time of sampling in January of 2001 were considered. Demographic and placement variables were controlled for. Health and disability variables included mental health (clinical score on the CBCL as administered to caregivers), health (ICD-09 diagnosis based on nurse reviews of case records, not inclusive of mental retardation), and disability (delivery of special education services for reasons other than emotional or behavioral disorders). Cox proportional hazards models were fitted using relevant predictor variables to model impact of predictors on time to reunification, adoption, or third placement move. Results: Findings indicated that child characteristics had only minimal impact on either reunification or placement stability. When all factors were considered, only age at entry was predictive of either outcome (hazards ratios for age at entry were 1.08 and 1.10 respectively). However, child characteristics were strongly associated with adoption outcomes. Placement in group care was removed from the model due to the fact that no child in group care was adopted. Age at case opening was again a significant predictor (hazards ratio 0.89). However, the only other significant predictor was a clinical-level CBCL score (hazards ratio 0.45). Physical health and disability were not significant predictors of likelihood of adoption. Implications: Children with physical health conditions and developmental disabilities are often stereotyped as difficult to adopt. These results indicate that these conditions do not pose a significant barrier to adoption while the impact of emotional and behavioral characteristics is profound. The implication is that children in foster care should be assessed early for mental health concerns and that appropriate services should be put in place and maintained both in order to maximize child well-being and to optimize opportunities for permanent placement.