In 1997, the New Mexico Children, Youth and Families Department (CYFD) and CRC developed a new child protective services case management system. To support this effort, CRC and CYFD staff conducted a risk assessment study of 1,450 sample families investigated for abuse or neglect during 1994-95.  Based on the findings of this research effort, CPS staff in New Mexico decided to employ risk assessment procedures at investigation for both substantiated and unsubstantiated families. In addition, they determined that high risk, unsubstantiated cases should be systematically referred to voluntary services for preventive intervention. This report examines study findings that led to the administrative decision to screen unsubstantiated families with higher levels of risk and service needs into prevention services. New Mexico’s experience provides an example of how service delivery planning may be improved by employing empirical procedures to examine case outcomes within the context of a risk assessment study.