News Release

Study finds family members are most common perpetrators of infant and child homicides in the U.S.

Analysis of 44 years of FBI data sheds light on possible prevention strategies for pediatric homicide

Peer-Reviewed Publication

Texas A&M University

Homicide is a leading cause of death among people 19 and younger in the United States, and firearms are used in a large proportion of these crimes. Although the homicide rate for this population declined in the 1990s and 2000s, it has increased every year since 2013, with a large spike during the COVID-19 pandemic.

Now, new research by Hannah Rochford, PhD, an assistant professor with the Texas A&M University School of Public Health, and two colleagues from the University of Iowa, provides insight into the characteristics of those who committed these crimes and their use of firearms from 1976 to 2020.

“The more national-level information we have about these perpetrators, the better we can develop comprehensive, evidence-based public health policies and prevention strategies,” Rochford said. “Unfortunately, the data has lagged behind data that is known for most other public health challenges. For example, the National Violent Death Reporting System did not exist at all before 2003, did not include a majority of states until 2015, and still does not fully represent violent deaths in all states. This makes it difficult to learn from past trends, like the youth violence surge of the late 1980s and early 1990s.”

For their study, published in Injury Epidemiology, the researchers sought to fill these gaps by describing trends between 1976-2020 in perpetrator characteristics (sex, age and relationship to victim) and firearm presence by the age, sex, and race of U.S. homicide victims from birth to 19 years of age.

To do so, researchers applied the multiply-imputed version of the Federal Bureau of Investigation’s 1976–2020 Supplementary Homicide Reports (SHR).

“The unimputed SHR is limited by ‘unit missingness’ because not every law enforcement agency provides data for the reports every year, and ‘item missingness’ because some homicide incidents were missing case information,” Rochford said. “For example, more than one quarter of homicide incidents were missing information on the perpetrator’s age, sex and race.”

After stratifying the descriptive analyses by victim age group, sex, race and five-year time periods, the team found that family members were the most common perpetrator of infant and toddler (ages 0–4) and child (ages 5–12) homicides, and acquaintances accounted for the majority of adolescent (ages 13–19) homicides. One quarter of adolescent homicides with female victims were perpetrated by an intimate partner.

The team found there was a sustained increase in the proportion of homicides committed with a firearm. From 2016 to 2020, the proportion of homicides that involved firearms was an all-time high across the study period for three categories: infants and toddlers at 14.8 percent, children at 53.1 percent, and adolescents at 88.5 percent. Firearm homicide was particularly burdensome to Black pediatric victims, with Black infants and toddlers experiencing twice the burden as White infants and toddlers, for example.

“These differences appear to align with developmental changes in family dependency and interaction, peer and romantic relations, and age-related role independence,” Rochford said. “For example, adult female family members were responsible for more than a quarter of all infant and toddler homicides, but to less than 1 percent of adolescent homicides.”

Rochford said these findings indicate that policy interventions that improve family stability and well-being may be most effective at preventing infant, toddler and child homicides, and programs that target peer and community relationships, as well as policies that focus on firearm access, may be more crucial for preventing adolescent homicides.

By Ann Kellett, Texas A&M University School of Public Health

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.