Child maltreatment fatalities in children under 5: Findings from the National Violence Death Reporting System

Abstract

Objective

To describe the distribution of child maltreatment fatalities of children under 5 by age, sex, race/ethnicity, type of maltreatment, and relationship to alleged perpetrator using data from the National Violent Death Reporting System (NVDRS).

Study design

Two independent coders reviewed information from death certificates, medical examiner and police reports corresponding to all deaths in children less than 5 years of age reported to NVDRS in 16 states.

Results

Of the 1,374 deaths for children under 5 reported to NVDRS, 600 were considered attributable to child maltreatment. Over a half of the 600 victims of child maltreatment in this age group were under 1 year old, 59% were male, 42% non-Hispanic Whites, and 38% were non-Hispanic Blacks. Two thirds of child maltreatment fatalities in children under 5 were classified as being due to abusive head trauma (AHT), 27.5% as other types of physical abuse, and 10% as neglect. Based on these data, fathers or their substitutes were significantly more likely than mothers to be identified as alleged perpetrators for AHT and other types of physical abuse, while mothers were more likely to be assigned responsibility for neglect.

Conclusions

Among children under 5 years, children under 1 are the main age group contributing to child maltreatment fatalities in the NVDRS. AHT is the main cause of death in these data. These findings are limited by underascertainment of cases and fair inter-rater reliability of coding.

Practice implications

The findings suggest the need to develop and evaluate interventions targeting AHT to reduce the overall number of child maltreatment deaths in young children. These interventions should make special efforts to include fathers and their substitutes.

Introduction

Homicide is the fifth leading cause of death among children under 5 years of age (National Center for Injury Prevention and Control, 2006) and almost half of homicides in young children are attributed to child maltreatment (Bennett et al., 2006). Children in this age group consistently account for more than 80% of the fatal cases of child maltreatment (USDHHS, 2005, USDHHS, 2006, USDHHS, 2007, USDHHS, 2008). Information on the characteristics of these deaths is needed to guide public health services as to what sectors of the population and types of maltreatment should be targeted for prevention programs and research.

Existing national data on child maltreatment fatalities are limited to the data reported by the National Child Abuse and Neglect Data System (NCANDS). These data are based on reports to the Children's Bureau from states’ Child Protective Services agencies (CPS). The number of child maltreatment fatalities based on CPS reports is believed to be underestimated. This has been attributed to the difficulties of determining whether death was intentional, the cause of death being determined by a different agency that does not report to CPS, and the differing methodologies and definitions used by agencies that report to CPS (Crume et al., 2002, Hermans-Giddens et al., 1999, McClain et al., 1993). Although CPS in some states obtain additional information from other agencies such as their health departments, vital statistics departments, medical examiners offices, and fatality review teams, this is not uniform across states. In addition, the data reported often do not provide sufficient detail on perpetrators’ characteristics or circumstances of the event to inform policy decisions or the development of prevention programs for specific types of maltreatment.

In 2002, the Centers for Disease Control and Prevention developed the National Violent Death Reporting System (NVDRS) to provide detailed data on violent deaths among children and adults (Steenkamp et al., 2006). NVDRS is an active, state-based surveillance system that collects information on homicides, suicides, unintentional firearm deaths, deaths from legal intervention (e.g., involving a person killed by an on-duty police officer), and deaths of undetermined intent such as those for which there is insufficient information to enable a medical or legal authority to make a distinction among unintentional injury, self-harm, or assault. Seventeen states currently contribute data to NVDRS: Alaska, Maryland, Massachusetts, New Jersey, Oregon, South Carolina, and Virginia since 2003; Colorado, Georgia, North Carolina, Oklahoma, Rhode Island, and Wisconsin since 2004; and California (four counties), Kentucky, New Mexico, and Utah since 2005.

NVDRS is different from other surveillance systems in that it collects and links data from multiple existing data systems, specifically, death certificate data from health departments and vital statistics departments, coroner/medical examiners’ reports, and police reports on circumstances and alleged perpetrator. Thirteen states collect data from supplementary homicide reports when available; nine states sometimes receive information from hospitals or emergency departments; and eight states occasionally receive information from child fatality review reports. Sociodemographic information for victims and alleged perpetrators is collected, as well as information on the relationship of the perpetrator to the victim, and circumstances and precipitating factors contributing to deaths. Data are added to the system from all sources as they become available with a previously assigned data source hierarchy for use when there are inconsistencies across data sources. NVDRS then links these data from the different sources to provide a more comprehensive view of violent deaths. Given the multisource nature of NVDRS data it is possible that this surveillance system may be able to provide a more comprehensive picture of child maltreatment fatalities.

The purpose of this study was to establish the distribution of child maltreatment fatalities in children under 5 captured by NVDRS by age, sex, race/ethnicity, type of maltreatment, and relationship to alleged perpetrator. This information will help guide public health efforts in setting priorities for prevention research and intervention.