Objective: This descriptive study examines 45 child and adolescent fatalities related to restraints in residential (institutional) placements in the United States from 1993 to 2003. Method: The study team used common Internet search engines as its primary case discovery strategy to determine the frequency and the nature of the fatalities, as well as the characteristics of the children and the adolescents involved. Results: Male children and adolescents were over-represented in the study sample. Thirty-eight of the fatalities occurred during or after a physical restraint, and 7 fatalities occurred during the use of mechanical restraints. Twenty-eight of the deaths occurred in a prone restraint. In 25 of the fatalities, asphyxia was the cause of death. Conclusion: In the 23 cases in this study where information is available, none of the child behaviors or conditions that prompted the restraint would meet the standard of danger to self or others: the commonly accepted criteria for the use of a restraint. The study points to deficiencies in fatality reporting, recommends reporting fatalities to established state child fatality review boards, and reinforces that restraints be governed by strict protocol and monitoring. The study also urges caution to policymakers in substituting or changing restraint procedures based on the incomplete data reported in this study.