Addressing over reporting of child abuse
Pilot project looks to tackle racial bias
A pilot project aimed at lowering the rate of medical professionals who overreport Black, Latino and Indigenous families to the child welfare system is launching in Michigan, the American Bar Association announced.
The community-based initiatives in Detroit and Ann Arbor hope to address racial bias and inequity the association says make medical professionals significantly more likely to report injuries to Black children as child abuse.
While 31% of kids in Michigan are children of color, they make up more than half of the state’s foster care population.
Failing to identify cases of physical or sexual abuse of children can have grave and even fatal consequences. But overreporting can have grievous outcomes too, especially when unnecessary investigations end up entangling families in the court system and pushing kids into the child welfare system.
“Sometimes children are not safe, and that is when child protection needs to step in — when children aren’t safe. It is equally important to remember that removing children from their parents is a last resort because of the trauma associated with being removed from one’s family,” said Dr. Ann Stacks, director of the Infant Mental Health Program at the Merrill Palmer Skillman Institute at Wayne State University and part of Michigan’s Stop Overreporting Our People (STOP) initiative. “It is so important to make sure that children aren’t being unnecessarily investigated or removed.”
Demetrius Starling is senior deputy director of children’s services administration for the Michigan Department of Health and Human Services, and one of three state leaders selected for the STOP project’s national advisory board.
“We want to keep children safe by recognizing those biases within our system, and also to really take a larger look about the overrepresentation of children of color in the foster care system that can start with some of these overreporting instances,” Starling said.
National data show bias in reporting children of color
In Michigan, Starling said, the preponderance of child welfare reports come from the medical community, law enforcement, schools and the courts. A deeper dive showed that children of color and those from tribal communities were being reported disproportionately often.
That aligns with nationwide findings. A 2022 study found that “Black patients were disproportionately overrepresented, composing 12% of the U.S. population, but 33% of suspected child abuse patients.” Other research has noted significant race and ethnicity-based disparities in how physicians evaluate and report abusive head trauma.
According to a 2021 study, “Once reported, cases with Black children are more likely to be accepted for investigation, be confirmed, be brought to court, result in removal of the children from their families for longer periods of time, and take longer to be closed, possibly related to surveillance bias.
“Multiple points in this process are subject to bias,” the study authors wrote. “But the process begins with reporting.”
A recent analysis of Michigan child welfare data revealed Black children were twice as likely as white children to be the subject of a maltreatment investigation, but had nearly identical removal rates following investigations. In Kent County, virtually all disproportionality in the foster care system was introduced at the initial point of child welfare contact, the researchers found.
Michigan investigates approximately 69,000 cases of child abuse every year. Around 25,000 of them are confirmed to be victims of maltreatment. Michigan’s rate of child maltreatment consistently surpasses the national rate.
Gut feelings and intuition
Michigan’s project proposal was one of eight selected by the American Bar Association’s Center on Children and the Law, which received funding from the U.S. Department of Health and Human Services to address race equity in the country’s child welfare systems.
The Michigan project will launch in Detroit and Ann Arbor under the direction of an advisory board and will apply for renewal for up to five years.
The pilot’s first phase will include efforts to capture more data that could help officials understand where medical overreporting is more prevalent and by whom. Focus groups and a root cause analysis will then help tackle a harder question: Why?
Starling fingered a lack of cultural awareness and training. A family might bring a child to the emergency room or a clinic if they don’t have good access to a pediatrician they can visit during work hours — something a medical professional could misinterpret as neglect. Medical professionals need to ask the right questions and evaluate the situation in front of them before jumping to conclusions, Starling said.
“Because caregivers rarely admit to injuring their children, such reports rely in part on providers’ gut feelings, making them susceptible to unconscious, systemic bias,” according to an article detailing research on the topic by Stanford pediatric surgeon Dr. Stephanie Chao.
Making snap judgments works in most high-stakes medical situations where decisions must be made quickly, said Dr. Ann Botash, a co-author of the 2021 study who is also a professor of pediatrics and director of the child abuse referral and Evaluation Program at SUNY Upstate Medical University.
Botash called this System 1 thinking, described as intuitive, automatic and based on pattern recognition by using rules of thumb and mental shortcuts.
Since using that type of thinking to diagnose child abuse relies on intuition and experience, it could make that type of error hard to overcome, she wrote.
“You have an impression, a gut feeling, you make a report,” Botash said. “And I don’t think we should stop doing that because most of the time we are right. But we should stop once we think that and say, ‘Why? Why am I thinking that?’ And make sure that it isn’t rooted in my own bias.”
Mandated reporters are told it’s not their job to figure out whether abuse or neglect exists, but to report their suspicions and let Child Protective Services figure it out, said Stacks. Michigan does not require training for mandated reporters, though information and training resources are available on various websites.
Interventions to help check bias
The designs for Michigan’s pilot projects have not been drafted, much less finalized. That’s intentional. The first phase of the project will focus on evaluating the need and thinking through possible interventions.
“I think this project is unique because a multidisciplinary group of experts is working to develop cross-system solutions to a longstanding issue in child welfare,” Stacks said. She lauded Starling for his use of data-driven decision-making and commitment to addressing racial disparities systemwide.
Starling said working groups created to guide these interventions will pull heavily from affected communities, including tribal government partners, medical professionals who have made CPS reports and families who ended up involved with the child welfare system after interacting with the medical system.
“When you engage with disenfranchised communities that have been heavily impacted by our policies, our procedures, our statutes, sometimes you can’t just go in like a bull in a china shop and say, ‘This is what we want to do,’” Starling said. “We’re kind of unraveling some of those issues or concerns that have plagued our communities for years.”
One solution could be screening tools medical professionals could use to dispassionately evaluate possible child abuse cases based on the evidence at hand.
Starling pointed to training and technical assistance in the form of oversight and guidance as tools that had been effective in other state initiatives that engaged the private sector. A possible model could be based on the Residential Collaboration and Technical Assistance Unit, which provides on-site services and support to child caring institutions. That unit was launched last spring as part of Michigan’s Keep Kids Safe Action Agenda.
Other interventions could include beefing up multidisciplinary teams in hospitals or advocacy centers that could serve as bias checks before a report is submitted. Or designing a universal screening tool as part of a patient’s electronic medical record that would help medical professionals analyze kids’ injuries indiscriminately.
Any of these could help address the other side of the racial bias problem: the fact that injuries to white children are often underreported as suspected abuse.
“I think this project is unique because a multidisciplinary group of experts is working to develop cross-system solutions to a longstanding issue in child welfare.”
Dr. Ann Stacks
Director of the Infant Mental Health Program at the Merrill Palmer Skillman Institute at Wayne State University and part of Michigan’s Stop Overreporting Our People (STOP) initiative
Healing and expanding
Starling hopes that having those voices at the table will ensure whatever interventions are piloted in Michigan can help address longstanding issues, including mistrust by communities of color in the medical system. The final phase of the project has community and racial healing as its goal.
“It’s meant to repair the hurt done to communities of color and reduce any potential further harm caused by our system,” Starling said. “So that’s a big part of this. And it’s going to probably take quite a bit of time for us to kind of really synthesize that information and make sure we’re not tokenizing their voices.”
Michigan has its eye on other communities of mandated reporters outside of medical professionals for future initiatives that could break down bias.
“We have talked about overreporting and disproportionality and biases within our systems for a long time,” Starling said, “so to actually have federal support for this with the monies and resources is truly amazing.”