Using Big Data to Improve Systems That Help Children
Big problems call for big tools. University of North Carolina at Chapel Hill School of Social Work Assistant Professor Rebecca Rebbe uses big data from big systems — including data from California’s Child Protective Services and Rady Children’s Hospital in San Diego — to discover compelling levers for policy reforms that better protect children from maltreatment.
Rebbe received an R21 grant from the National Institute of Child Health and Human Development in 2021 that provided direct cost funding of $275,000 over two years. The grant funded her research study that looks at whether child maltreatment increased during the COVID-19 lockdowns when children had less interaction with teachers, health care providers and others who typically notice and report suspected maltreatment to child protective services (CPS), which can then intervene.
As the principal investigator on the study titled “The Impact of COVID-19 on Child Maltreatment-related Medical Encounters and System Responses Using Linked Administrative Data,” Rebbe compared hospital records on children seen for suspected maltreatment with information reported to CPS during that same time period. Rebbe wanted to know whether the stresses of the pandemic and interruption of services resulted in more harm to children.
“At the beginning of the pandemic, there was a lot of media attention on what would happen with kids if they’re not in schools and families aren’t getting services they need,” Rebbe said. “There were a lot of concerns, but not a lot of data.”
At the start of the pandemic, the number of reports of suspected maltreatment submitted to CPS dropped. Whether the drop was due to a lack of interaction between children and reporting authorities or fewer concerns of maltreatment, Rebbe had no way of knowing. But CPS reports, a traditional data source of risk of harm to children, were suddenly vulnerable and rendered a weak source, she said.
From a prior study, Rebbe had a relationship already with Rady Children’s Hospital. That facilitated a partnership that allowed Rebe to access data on medical encounters (hospital admissions and emergency room records) before and after the height of the pandemic for her R21 study.
From that information, she could see whether child maltreatment related to injuries changed by number and type. By linking it to CPS data, which recorded a child’s history with the agency and trajectory, Rebbe could gain insight into how children were moving between systems.
The data showed that reports of children being hospitalized for maltreatment rose immediately with the onset of COVID-19, then dropped back to typical levels. Professionals have some discretion on what and when to report.
Though Rebbe didn’t have information on why reporting decisions were made, her hypothesis was that the child maltreatment concern would go undetected because children weren’t in school and interacting in person with other professionals, which drove medical personnel to err on the side of reporting.
The questions I ask, the variables I use, the data — it’s integral to the work I do, and what I saw or didn’t see on the ground. That’s definitely important to strengthen the systems that serve children and families.
REBECCA REBBE, UNC SCHOOL OF SOCIAL WORK ASSISTANT PROFESSOR
By integrating the data, Rebbe “de-siloed” the systems to gain a greater understanding of how children move between the medical and social services systems.
Early in her career, Rebbe was a social worker at a state agency, intervening with children and families. She shifted her focus to big data and systemic factors because she wanted to understand whether the cases she saw were typical or outliers.
“I was plagued by questions of what’s happening at a broader level,” she said. “How was the system functioning? Are we improving the lives for kids and families? Where can improvements be made?”
Rebbe wants to continue to look more broadly at histories and trajectories of maltreated children and understand longitudinally how they are faring in both medical and social services systems, how these systems work together and where there are gaps or obstacles. Data science techniques are tools to understand what improvements to systems need to be made to benefit children’s lives.
“I try to make sure I balance my perspective as a researcher with my experience as a practitioner, but my practice experience shapes everything I do,” Rebbe said. “The questions I ask, the variables I use, the data — it’s integral to the work I do, and what I saw or didn’t see on the ground. That’s definitely important to strengthen the systems that serve children and families.”
Advancing equity. Transforming systems. Improving lives.
UNC School of Social Work
The University of North Carolina at Chapel Hill
Tate-Turner-Kuralt Building
325 Pittsboro Street | Campus Box 3550
Chapel Hill, NC 27599-3550