Comparing fatal child abuse involving biological and surrogate parents

Abstract

Background: Nearly half of pediatric homicides younger than 5 years are attributable to child abuse. Parents are most commonly the perpetrators, but less is known about incidents involving biological versus surrogate parents. We sought to evaluate the characteristics of fatal child abuse involving biological and surrogate parents using the Georgia National Violent Death Reporting System, which we believe may differ in demographics and incident characteristics.

Methods: This database was used to examine all homicides of children younger than 18 years from 2011 to 2017. Demographics and incident characteristics were analyzed using the existing National Violent Death Reporting System variables and incident narratives. Nonparametric and χ2 tests were used to compare fatal child abuse incidents involving biological and surrogate parents (e.g., adoptive, foster, step-parents, intimate partners of biological parent).

Results: There were 452 pediatric homicides and 219 cases of fatal child abuse. Of all cases of fatal child abuse, 60% involved biological and 29% involved surrogate parents. Compared with children killed by biological parents, children killed by surrogate parents were older (4 vs. 3 years), more often male (71% vs. 51%), more likely to survive the initial injury and present to the emergency department before death (96% vs. 69%), and less likely to have a medical comorbidity (2% vs. 11%; all p < 0.05). Surrogate parents were more likely to be male (90% vs. 48%) and use a firearm (20% vs. 13%) to inflict the injury, both p < 0.05. The race/ethnicity of the child was not associated with the parent's relationship.

Conclusion: Child abuse accounts for half of all pediatric homicides. Parents are the most common perpetrators of fatal child abuse, but surrogate parent perpetrators are almost exclusively male and more likely to use firearms. Most children have a history of abuse, leaving an opportunity to intervene on potentially preventable deaths if abuse is identified in a timely fashion.

Level of evidence: Prognostic and Epidemiologic, Level III.