Review and special articleChild death review: The state of the nation
Introduction
I n 1998, over 19,000 children aged 1 to 18 years died in the United States.1 Twenty-six percent of these child fatalities were due to natural causes, while 74% resulted from injuries. Approximately 30% of these injury deaths were classified as intentional injuries (homicide, suicide) and 70% were classified as unintentional (accidents). Furthermore, estimates of annual child maltreatment (or child abuse and neglect) fatalities among children aged <18 years ranged from 1000 to 2600.2 While these statistics provide an overview of the nature and magnitude of the problem of child death in the United States, both the accuracy and the level of detail provided by current data sources are inadequate for successful prevention.3, 4, 5, 6
The evidence regarding the limitations of current data sources for accurately documenting the cause and manner of death among children originates primarily from the child maltreatment literature. Accurate identification and description of fatal child maltreatment are challenging for several reasons: the ease with which child maltreatment can be concealed, inadequate investigations, a lack of information sharing among agencies, a lack of definition consensus, and the nature of International Classification of Diseases coding requirements.4, 6, 7 Based on studies conducted in Missouri and North Carolina, current available data sources (death certificates, child protective services, and law enforcement records) clearly underestimate the incidence of child maltreatment fatalities when used in isolation.6
Recognition of the inadequacy of current sources for accurately identifying the causes of unexpected death among children led to the development of child death review (CDR) programs. Generally, CDR is a multidisciplinary, multi-agency process designed to examine the causes and circumstances of child deaths. This process is based on the premise that maltreatment is more difficult to conceal and less likely to be missed when professionals from various agencies and disciplines share information regarding the child and the circumstances of death. For example, autopsy findings indicative of child abuse, which were unavailable to the officers doing the death scene investigation, may shed new light on scene findings that were previously dismissed as insignificant. Inconsistent histories of the injury mechanism or discovery of the child may be revealed, highlighting a discrepancy between documented injuries and reported mechanism. Although originally developed to better identify maltreatment-related deaths, CDR is often used to facilitate prevention of child deaths more broadly, particularly injury-related deaths. Therefore, multi-agency multidisciplinary review has the potential to decrease misclassification of deaths, increase opportunities for effective intervention on behalf of surviving children, and prevent future deaths.4
The concept of conducting reviews of individual deaths dates back over 60 years with maternal mortality reviews. Maternal mortality reviews are conducted by multidisciplinary teams on the state level for the purpose of reducing pregnancy-related maternal mortality. Fetal and infant mortality review, started in 1988 with the goal of reducing infant mortality, consists of local in-depth review of selected fetal and infant deaths using nationally standardized review criteria. The first multidisciplinary multi-agency CDR team was formed in Los Angeles County in 1978.7 Unlike fetal and infant mortality review, CDR aims to review all deaths (or all deaths due to certain causes), and the criteria for review are set at the state or local level. Although multiple organizations have endorsed CDR, including the American Academy of Pediatrics8 and the American Bar Association,9 there are no standardized criteria for CDR and no national guidance to CDR programs.
In 1991, the U.S. Public Health Service (USPHS) endorsed CDR.10 The 1998–1999 annual progress review of the Healthy People 2000 goals reported that 47 states and the District of Columbia had a CDR program.11 Unfortunately, the USPHS report contained only a program count without additional information on scope or process. Based on personal experience with CDR programs in several states, we postulated that CDR programs were highly variable by state. However, after searching for a summary of state programs, no comprehensive summary of program attributes was found to exist. Documenting program variability is important because the lack of standardization will affect efforts to accurately describe the causes and circumstances of death among children at a national or regional level. In order to compare information across states or perform national level surveillance of child deaths using CDR data, states must be conducting reviews in a comparable way and collecting data using standardized definitions. This study was conducted in an effort to describe the current status of CDR in the United States and to assess variability in program purpose, scope, organization, and process.
Section snippets
Study population
A representative from the CDR program in each state and the District of Columbia was identified using the contact list maintained by the Interagency Council on Child Abuse and Neglect–National Child Fatality Review (J. Langstaff; Interagency Council on Child Abuse and Neglect–National Child Fatality Review, personal/written communication, 2000). An attempt was made to contact the person identified on the list as a “primary contact” in each state. When this primary contact could not be reached,
Results
A 100% response rate was achieved. CDR representatives in all 50 states and the District of Columbia (hereafter referred to as a state for convenience) were contacted, agreed to participate, and returned the mail questionnaire. Two states reported that they currently had no program or CDR process. Follow-up phone interviews were completed for each state with a program. Results presented here are for the 49 states with active CDR at the time of data collection. Table 1 displays several key
Discussion
This report is the first systematic assessment of CDR in the United States since 1992.7 The 100% response rate provides a solid foundation for this comprehensive assessment and description of the current status of CDR nationwide. Although detailed description of the intricacies of the CDR programs was impeded by the limited flexibility of the quantitative survey instruments to describe complex and diverse processes, this study provides important information on the progress in CDR over the past
Acknowledgements
We are grateful to Paul Matz, MD, Roderick King, MD, MPH, and Dalia Batista, MD, for their assistance in study design and conceptual input; Patrick Vivier, MD, for assistance with design and analysis; and Scott Laney for assistance with data collection. Special thanks to Lloyd Potter, PhD, and Stephen Wirtz, PhD, for reviewing an earlier draft of this manuscript. PGS was funded in part by a Career Development Award from the National Institute of Child Health and Human Development
References (14)
- National Center for Health Statistics. 10 Leading causes of death, United States: 1998, all races, both sexes. Atlanta...
- et al.
Estimates of fatal child abuse and neglect, United States, 1979 through 1988
Pediatrics
(1993) - U.S. Advisory Board on Child Abuse and Neglect. A nation’s shame: fatal child abuse and neglect in the United States....
- et al.
Underascertainment of child abuse mortality in the United States
JAMA
(1999) - National Clearinghouse on Child Abuse and Neglect Information. Highlights from child maltreatment 1999. Washington DC:...
- et al.
The Missouri Child Fatality Studyunderreporting of maltreatment fatalities among children younger than five years of age, 1983 through 1986
Pediatrics
(1993) - et al.
Origins and clinical relevance of child death review teams
JAMA
(1992)