Date of Award

Summer 8-7-2018

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Public Health

First Advisor

Shannon Self-Brown

Second Advisor

Beverly Fortson

Third Advisor

Derek Ford


Background: Homicide is the fifth leading cause of death among children from birth to 17-years-old, with approximately 1,700 child maltreatment (CM)-related homicides occurring in the U.S. annually. In 2016, more than three-quarters (78%) of these deaths involved biological parents acting alone, together, or with other individuals, and approximately 17% were perpetrated by a nonparent, suggesting different victim-perpetrator relationships present different levels of risks. The present study examined the association between child, family, and perpetrator characteristics and method of lethality used in CM-related homicides in the context of victim-perpetrator relationship. Methodology: Data are from the National Violent Death Reporting System (NVDRS). NVDRS captures data using death certificates, law enforcement (LE) reports, and coroner/medical examiner (C/ME) reports. Using content analysis, which is a research technique to systematically code textual material into categorical data, CM-related homicides of children ages 0-17 for 2012-2015 in 32 states were examined. Bivariate and multinomial logistic regression models were used to evaluate associations between method of lethality used in CM-related homicides and victim-perpetrator relationship, and child, family, and perpetrator characteristics. Results: During the 2012-2015 data collection period, 996 children were victims of CM-related homicide. Biological fathers were the most common perpetrators (37.8%), followed by mother’s male companion (26.8%), biological mother (21.8%), and “other” perpetrator (13.6%). With respect to method of lethality, more than one third of the children were beaten/bludgeoned to death (37.3%) and deaths by “other” means was the second most prevalent method of lethality (24.1%). Further, the odds of a child being beaten/bludgeoned to death versus dying by abusive head trauma (AHT) among those killed by mother’s male companion was 1.98 (95% CI [1.02, 3.88]) times greater the odds of being beaten/bludgeoned to death by biological fathers, adjusting for all other predictors in the model. Moreover, the presence of a bystander significantly increased the odds of a child being beaten/bludgeoned to death, Adjusted Odds Ratio (aOR) = 2.31, 95% CI [1.04, 5.14]. In addition, the presence of intimate partner violence, parental relationship conflict, and arguments were each associated with increased odds of firearm-related deaths in children versus death by AHT, aOR = 8.67, 95% CI [2.60, 28.91], aOR = 9.17, 95% CI [1.78, 47.18], and aOR = 13.85, 95% CI [2.51, 76.52], respectively. Conclusion: This study helps to better understand the circumstances and characteristics of CM-related homicides, which may inform primary prevention efforts, prevent child death, and, when used in the context of a comprehensive prevention strategy, may help in assuring safe, stable, nurturing relationships and environments for all children.