“The central question of this research was: does child maltreatment lead to behavior problems or vice versa?” said Shenk, who is also a Social Science Research Institute co-funded faculty member through the Child Maltreatment Solutions Network. “Short term or cross-sectional studies can’t capture this information, so we combined a unique dataset with a novel statistical approach to show that children who were maltreated were more likely to display subsequent behavior problems at multiple ages throughout childhood and adolescence.”
The researchers used data from the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN), which were collected every two years from when children were four until they turned 18. Using seven waves of data from 1,354 children and their caregivers between the ages of four and 16, the researchers identified whether the children had a documented experience of maltreatment. Then, they simultaneously examined the relationship between maltreatment exposure and externalizing and internalizing behaviors.
The researchers employed a random-intercept, cross-lagged panel model, which they said helped them understand how exposure to maltreatment, or changes in internalizing and externalizing behaviors, influence one another over time. At the majority of timepoints throughout childhood and adolescence, children who had experienced maltreatment displayed an otherwise unexplained increase in behavior problems during the next wave of data collection.
When looking at the data in the other direction — examining whether children who displayed behavior problems were more likely to be maltreated within the next two years — they found that children were generally not more likely to be maltreated if they displayed increases in problem behaviors. The one exception to this trend was that children who displayed increases in externalizing behaviors at age 10, like noncompliance or delinquency, were more likely to experience maltreatment two years later.
Olson said this could be a significant finding for understanding how abuse and behavior problems are related.
“The ages of 10 to 12 are when children start to transition to adolescence,” Olson said. “As this happens, peers become more important to them, and individuals begin to think more for themselves, which can result in more externalizing behaviors. This is the only period during the study when an increase in externalizing behavior problems led to increased maltreatment. At the same time, exposure to maltreatment during this period also led to an increase in behavior problems.”
The researchers said that modeling this two-way relationship between behavior problems and maltreatment simultaneously may represent a unique opportunity for enhanced screening and the type and timing of interventions.
“Screening for maltreatment exposure and behavior problems in pediatrician offices is an important take away from this study," Shenk said. "Screening during the ages of 10 to 12 is especially important given maltreatment and externalizing behaviors were related to one another at this point in development. There are brief, scientifically valid screenings that pediatricians can use for maltreatment and behavior problems. Additionally, when pediatricians identify children who have experienced maltreatment or demonstrated increased behavior problems, there are interventions that can work to decrease both.”
The researchers pointed to parent-child interaction therapy — which was developed to address externalizing behaviors in a family context — as an excellent potential tool for reducing problem behaviors prior to age 10. They said the therapy teaches children to manage their behavior and often results in significant reductions in behavior problems. It has also been shown to prevent future abuse because caregivers learn new behaviors and strategies for promoting family relationships and addressing externalizing behaviors without resorting to violent words or actions.
Shenk, who has a joint appointment in the Penn State College of Medicine Department of Pediatrics, expressed excitement about the potential of this research to guide better evaluation and intervention by pediatricians.
Olson said she was surprised at how robust and clear the results were.
“Everyone in Dr. Shenk’s lab wants to ask questions that will move the needle on our understanding of child maltreatment,” Olson explained. “On this study, we found that child maltreatment robustly predicts subsequent behavior problems at multiple ages, but we also learned that there is little evidence that behavior problems increase the risk for subsequent maltreatment.”
John Felt of the Penn State Center for Healthy Aging, Emily Dunning of the Penn State Department of Human Development and Family Studies, Zhenyu Zhang of the Penn State Department of Psychology, Metzli A. Lombera of the Penn State Department of Psychology, Camille Moeckel of the Penn State College of Medicine, Manal U. Mustafa of the Penn State College of Medicine's Department of Pediatrics, Brian Allen of the Penn State College of Medicine's Department of Pediatrics and Department of Psychiatry and Behavioral Health, and Lori Frasier of the Penn State College of Medicine Department of Pediatrics also contributed to this research.
The National Institutes of Health and the U.S. National Science Foundation funded this research.