Child abuse inflicted by caretakers has been reported the leading cause of morbidity and mortality in children. Although researchers have attempted to estimate the rate of child abuse homicide nationwide, its true scope is unknown. Quantifying child abuse homicide is important since this type of homicide is arguably preventable.

This study utilizes the National Violent Death Reporting System (NVDRS) to analyze child abuse homicides inflicted by the child's caretaker across 17 U.S. states during the years they participated from 2003 to 2005. The NVDRS reports violent death data from multiple official sources, providing this analysis with a more accurate number of child abuse homicides than previously available. In this thesis, I established if the NVDRS ascertains more abuse or non-abuse homicides than death certificates. I reviewed ascertained and misclassified deaths to assess whether a common set of victim or incident characteristics exist across these homicides. A logistic regression was also conducted to examine the probability of child abuse homicide given these characteristics.